CN108205768B - Database establishing method, data recommending device, equipment and storage medium - Google Patents

Abstract

The embodiment of the invention discloses a database establishing method, a data recommending method, a device, equipment and a storage medium. The database establishing method comprises the following steps: acquiring user historical data and commodity historical data from a data source; extracting user features according to user historical data and commodity historical data and preset user dimensions to form a user feature vector of a user, wherein the user feature vector comprises at least one commodity feature; according to user history data and commodity history data, extracting commodity features according to preset commodity dimensions to form a primary commodity feature vector of a commodity, wherein the primary commodity feature vector comprises at least one user feature; combining the characteristic vectors of the primary commodities according to a preset commodity association rule to form a characteristic vector of a secondary commodity; and storing the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector as a recommendation data set in a database. The invention can improve the accuracy of the prediction of the user attention information.

Description

Database establishing method, data recommending device, equipment and storage medium

Technical Field

The embodiment of the invention relates to a big data processing technology of a computer, in particular to a database establishing method, a data recommending method, a device, equipment and a storage medium.

Background

Under the condition that the online sales behavior is popularized, effective prediction on the behavior of purchasing commodities by users is a development trend.

In the prior art, the following behavior data is generally predicted based on the historical behavior data of the user. For example, a user often purchases clothes in the past, and it is predicted that the information on the product recommending clothes to the user is in accordance with the consumption behavior of the user. Besides the technology of counting and predicting the personal historical data of the user, the prior art generally adopts a correlation rule mode to recommend information for the user. Generally, mass user data is subjected to centralized statistics, and a behavior pattern of a user with certain attributes is determined. For example, if it is determined that a male user with an age range of 20-30 years may be interested in the electronic goods based on historical data of a large number of users, it is predicted that the electronic goods pushed to the user are in line with the consumption behavior of the user who meets the above characteristics.

However, in the course of research for carrying out the present invention, the inventors found that the above-described techniques had drawbacks in that: if the user's personal historical data is relied on, the range of the recommendation information is very limited, and the prediction result of the recommendation information is often not in line with the development trend of the user consumption behavior. Based on the statistics of massive user data, because the difference between users is large and the regularity is not obvious, if the statistical data size is not enough, the accuracy of predicting the consumption behavior of the user is low.

Disclosure of Invention

The embodiment of the invention provides a database establishing method, a data recommending device, equipment and a storage medium, and aims to improve accuracy of prediction of user attention information.

In a first aspect, an embodiment of the present invention provides a database establishment method, including:

acquiring user history data and commodity history data from a data source;

extracting user features according to user historical data and commodity historical data and preset user dimensions to form a user feature vector of a user, wherein the user feature vector comprises at least one commodity feature;

according to the user historical data and the commodity historical data, commodity features are extracted according to preset commodity dimensions to form a primary commodity feature vector of the commodity, wherein the primary commodity feature vector comprises at least one user feature;

combining the characteristic vectors of the primary commodities according to a preset commodity association rule to form a characteristic vector of a secondary commodity;

and storing the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector as a recommendation data set in a database.

In a second aspect, an embodiment of the present invention provides a data recommendation method, including:

acquiring an online recommendation requirement;

acquiring online data and recommendation rules from the online recommendation requirements;

and generating recommendation data based on the online data, the user characteristic vector in the database and the commodity characteristic vector by adopting the recommendation rule, wherein the user characteristic vector comprises at least one commodity characteristic, and the commodity characteristic vector comprises at least one user characteristic.

In a third aspect, an embodiment of the present invention provides a database creating apparatus, including:

the historical data acquisition module is used for acquiring user historical data and commodity historical data from a data source;

the commodity feature acquisition module is used for extracting user features according to user historical data and commodity historical data and preset user dimensions to form a user feature vector of a user, wherein the user feature vector comprises at least one commodity feature;

the system comprises a user characteristic acquisition module, a commodity dimension extraction module and a commodity dimension extraction module, wherein the user characteristic acquisition module is used for extracting commodity characteristics according to user historical data and commodity historical data and preset commodity dimensions to form a primary commodity characteristic vector of a commodity, and the primary commodity characteristic vector comprises at least one user characteristic;

the commodity feature processing module is used for merging the characteristic vectors of the primary commodities according to a preset commodity association rule to form a characteristic vector of a secondary commodity;

and the recommendation data forming module is used for storing the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector as a recommendation data set in the database.

In a fourth aspect, an embodiment of the present invention provides a data recommendation apparatus, including:

the demand acquisition module is used for acquiring online recommendation demands;

the information acquisition module is used for acquiring online data and recommendation rules from the online recommendation requirements;

and the information processing module is used for generating recommendation data based on online data, a user characteristic vector and a commodity characteristic vector in a database by adopting the recommendation rule, wherein the user characteristic vector comprises at least one commodity characteristic, and the commodity characteristic vector comprises at least one user characteristic.

In a fifth aspect, an embodiment of the present invention provides an apparatus, where the apparatus includes:

one or more processors;

a storage device for storing one or more programs,

when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the database building method provided by any embodiment of the present invention.

In a sixth aspect, an embodiment of the present invention provides an apparatus, where the apparatus includes:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement a database recommendation method provided by any of the embodiments of the invention.

In a seventh aspect, an embodiment of the present invention provides a storage medium containing computer-executable instructions, which when executed by a computer processor, are used to perform the database building method provided in any embodiment of the present invention.

In an eighth aspect, embodiments of the present invention provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the database recommendation method provided by any of the embodiments of the present invention.

The embodiment of the invention acquires the historical data of the user and the historical data of the commodity from the data source and extracts the characteristic vectors of the user and the commodity; according to a preset commodity association rule, commodity feature vectors are combined, user feature vectors, primary commodity feature vectors and secondary commodity feature vectors are stored, a recommendation database is obtained, commodity features are counted, and in addition to the features of people, the features of commodities are also considered in the recommendation process, so that the recommendation effect is more accurate.

Drawings

Fig. 1 is a schematic flowchart of a database establishment method according to an embodiment of the present invention;

fig. 2 is a schematic flowchart of a database establishment method according to a second embodiment of the present invention;

fig. 3 is a schematic flowchart of a data recommendation method according to a third embodiment of the present invention;

fig. 4 is a schematic flowchart of a data recommendation method according to a fourth embodiment of the present invention;

fig. 5 is a schematic flowchart of a data recommendation method according to a fifth embodiment of the present invention;

fig. 6 is a block diagram of a database creating apparatus according to a sixth embodiment of the present invention;

fig. 7 is a block diagram of a data recommendation device according to a seventh embodiment of the present invention;

fig. 8 is a block diagram of an apparatus according to an eighth embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.

Example one

Fig. 1 is a flowchart of a database establishing method according to an embodiment of the present invention. The method is suitable for the case of electronic commodity transaction, and can be executed by a database establishing device. The database building means may be implemented in software and/or hardware. As shown in fig. 1, the method includes:

and S110, acquiring user history data and commodity history data from a data source.

Acquiring offline data from a data source, and screening out needed user historical data and commodity historical data, wherein the offline data at least comprises the following steps: user behavior data, user attribute data, order data, and Stock Keeping Unit (SKU) data.

Further, the user history data may include user dynamic behavior data and user static attribute data; the user dynamic behavior data comprises at least one of a commodity browsing behavior, a navigation positioning behavior, a search engine searching behavior, a search result clicking behavior and a webpage browsing behavior; may be collected from an operation log of a user using various client software. The user static attribute data may include at least one of name, gender, address, height, weight, and school calendar, such as may be obtained from the user's registration information. The commodity history data may include commodity order data, commodity inventory data, and the like.

S120, extracting user features according to the user historical data and the commodity historical data and preset user dimensions to form a user feature vector of the user, wherein the user feature vector comprises at least one commodity feature.

Before extracting the feature vectors of the user historical data and the commodity historical data, the operations of removing the duplicate, removing the noise and the like are preferably carried out.

According to the user history data and the commodity history data, user features are extracted from a plurality of set dimensions according to preset user dimensions, wherein the dimensions can be any user attributes which are expected to be concerned, such as height, monthly consumption amount, the number and price of purchased mobile phones and the like. The dimension can be set manually or determined based on big data statistics by means of model training. And forming the user feature vector of each user by using the extracted user features. The user feature vector includes at least one commodity feature, for example, through the behavior of the user browsing a webpage, the commodity concerned by a certain user can be determined, and through the behavior of the user purchasing the commodity, the commodity preferred by the user can also be determined. The dimension characteristics of the commodities which the user is interested in can be used as commodity characteristics and added to the user characteristic vector. The commodity features may be features of a specific commodity, or may be abstract features describing the commodity, such as a replacement cycle of the electronic device, or a preferred color of a down jacket, or the like.

S130, according to the user history data and the commodity history data, commodity features are extracted according to preset commodity dimensions to form a primary commodity feature vector of the commodity, wherein the primary commodity feature vector comprises at least one user feature.

According to the user history data and the commodity history data, commodity features are extracted from a plurality of set dimensions according to preset user dimensions, wherein the dimensions can be any commodity attribute which is expected to be concerned, such as a brand to which the commodity belongs, monthly sales volume of each shop, a size, a color and the like. The dimension can be set manually or determined by big data statistics. The extracted commodity features are formed into a primary commodity feature vector for each commodity. The primary commodity feature vector comprises at least one user feature, which can be a user attribute suitable for the frequently-purchased crowd feature of the commodity. For example, a certain size of clothing, which may include suitable height characteristics, as user characteristics, etc.

And S140, merging the characteristic vectors of the primary commodities according to a preset commodity association rule to form a characteristic vector of a secondary commodity.

The commodity association rule is a relation established according to some same or similar parts among commodities, and can be freely set according to the attributes of the commodities or the behavior habits of users, some feature vectors of the primary commodities are combined according to the association rule of the preset commodities to obtain the feature vectors of the secondary commodities, for example, the feature vectors of the primary commodities such as color, model and material can be combined, the combination condition can be set according to specific needs, and the feature vectors of the secondary commodities are obtained after combination. The secondary commodity feature vectors can also be combined continuously based on the association rule to obtain new commodity feature vectors.

And S150, storing the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector as a recommendation data set in the database.

The storage method of each feature vector is not limited, and an array including various features may be generated and stored as independent feature vectors. Or establishing the relationship between various features to form a feature vector, and separately storing the content of the features and the relationship between the features.

The database establishing method provided by the embodiment of the invention comprises the steps of acquiring user historical data and commodity historical data from a data source, and extracting feature vectors of users and commodities; and combining the commodity feature vectors according to a preset commodity association rule, and storing the user feature vector, the primary commodity feature vector and the secondary commodity feature vector to obtain a recommendation database. By counting the commodity characteristics, the characteristics of people are considered in the recommendation process, and the characteristics of commodities are considered, so that the recommendation data are more accurate.

Example two

Fig. 2 is a flowchart illustrating a database establishing method according to a second embodiment of the present invention. The present embodiment further describes the system on the basis of the first embodiment. As shown in fig. 2, the method includes:

s210, obtaining user history data and commodity history data from a data source.

S220, extracting user features according to preset user dimensions according to the user history data and the commodity history data to form a user feature vector of the user, wherein the user feature vector comprises at least one commodity feature.

And S230, according to the user history data and the commodity history data, extracting commodity features according to preset commodity dimensions to form a primary commodity feature vector of the commodity, wherein the primary commodity feature vector comprises at least one user feature.

And S240, combining the primary commodity feature vectors of which the feature value similarity of the core features in the primary commodity feature vectors reaches the set condition according to at least one core feature set in the preset commodity association rule to form the secondary commodity feature vectors.

The core feature is an important feature of interest, and the primary product feature vectors of the respective products are generally combined based on the similarity of the set core features. The primary commodity feature vectors generally comprise core features and non-core features, and if one or more core features are the same or have similarity reaching a certain condition, the primary commodity feature vectors are combined. Wherein, a plurality of groups of secondary commodity feature vectors can be combined according to various different combination conditions. For example, when product color is a core feature, ten thousand sets of primary commodity feature vectors may be merged; when the product model is a core feature, only ten sets of primary commodity feature vectors may be merged. That is, the merging condition can be set according to specific needs.

The core characteristics can be set manually, and can also be screened and determined according to different recommendation requirements or big data statistics. Typical core features may include: the commodity self attribute, commodity sales data, commodity inventory data, commodity sales correlation degree and the like.

And S250, storing the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector as a recommendation data set in the database.

And S260, generating recommendation data according to the online recommendation requirement and the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector in the database.

And acquiring online recommendation requirements according to the online requirements of the users, acquiring commodity characteristics required by the users and the characteristic vectors of the users, screening the characteristic vectors of the primary and secondary commodities, and recommending commodities meeting the requirements of the users to the users.

And S270, correcting the characteristic values in the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector according to the online response data aiming at the recommendation data.

And according to the online demand data of the user, updating the characteristic vector of the user and the characteristic values in the characteristic vector of the primary commodity and the characteristic vector of the secondary commodity in real time, so as to ensure the accuracy of the data.

Further, the method may further include: and receiving the updated commodity association rule input by the administrator, and updating the characteristic values of the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector in the database according to the updated commodity association rule.

The above operation is equivalent to amending the database based on manual intervention data. Sources of human intervention data may include: advertisement data, trend data, right-adjusting data, random factors and the like, wherein the data is used as intervention factors to enter a database to influence data calculation results. Part of the commodity feature values need to be determined by statistics or prediction from offline historical data, and manual intervention data can modify the statistical algorithm or the prediction algorithm so as to adjust the calculated feature values. For example, for sales prediction of certain types of goods, it may be based on user search browsing data, actual purchase data, and popularity trend data, and the like, and the manual intervention data may adjust the weight among these types of considerations, so as to make the prediction result accurate.

For the adjustment of the feature vector feature value in the database, correction can also be performed based on a machine learning algorithm. The user feedback data and other feedback data are analyzed by a training learning technique of the computer to form correction data.

The commodity association rule may include at least one of: merging rules of the same type of commodities by taking the set commodity characteristics as classification bases; and taking the commodity characteristics with associated behaviors as commodity associated rules, wherein the associated behaviors comprise purchasing behaviors, using behaviors and the like.

The association rules of the goods are divided into static association rules and dynamic association rules, wherein the association rules based on specific attributes of the goods belong to the static association rules, for example: establishing association between two commodities when the similarity of set attributes such as resolution, CPU processing capacity, memory and the like of the mobile phone reaches set conditions, namely association behavior based on a static association rule; the dynamic association rule is an association rule based on the purchasing behavior of the user, such as: when a certain commodity is purchased and the probability of simultaneously purchasing another commodity reaches a set condition, establishing the association between the two commodities, which is the association behavior based on the dynamic association rule. The expression of the association can be various, for example, coarse-grained: the mobile phone is generally associated with the data line; fine particle size: the mobile phone with the model of millet and the mobile phone with the model of P9 are related; the association relationship may specifically include an association probability, for example, the probability that the mobile phone and the data line are purchased simultaneously is 70%.

According to the database establishing method provided by the embodiment of the invention, the recommendation data is generated according to the online recommendation requirement of the user, and the recommendation database can be continuously corrected according to the feedback data or intervention data from all aspects, so that the accuracy and the real-time performance of the data of the recommendation database are ensured.

EXAMPLE III

Fig. 3 is a schematic flowchart of a data recommendation method according to a third embodiment of the present invention. The embodiment is a data recommendation method based on the recommendation database of the embodiment, the method is suitable for data recommendation in commodity transaction, and the method can be executed by a data recommendation device. The data recommendation device may be implemented by means of software and/or hardware. As shown in fig. 3, the data recommendation method includes:

s310, acquiring online recommendation requirements.

The online recommendation requirement is a requirement which is generated in real time and needs to generate recommendation data based on the recommendation database, for example, when a user selects a certain commodity or generates an order, when a logistics service provider needs to provide stock forecast quantity and when the logistics service provider needs to actively recommend commodity information to the user who opens client software, the real-time online recommendation requirement is generated.

And S320, acquiring online data and recommendation rules from online recommendation requirements.

Further, online data are obtained from the online recommendation requirement, wherein the online data comprise at least one user characteristic and/or at least one commodity characteristic; and searching a corresponding recommendation rule according to the online recommendation requirement. The online data may include: user information, information concerning the commodity, and the like. The recommendation rule is a recommendation data calculation mode, a push mode and the like determined according to the condition of the online recommendation requirement.

S330, generating recommendation data by adopting a recommendation rule based on the online data, the user characteristic vector and the commodity characteristic vector in the database, wherein the user characteristic vector comprises at least one commodity characteristic, and the commodity characteristic vector comprises at least one user characteristic.

Searching related commodities in a data recommendation database and an online database according to the association rules of the user characteristic vector, the commodity characteristic vector and the commodities, generating recommendation data, and recommending the recommendation data to a demand subject, wherein the demand subject mainly comprises: users, logistics service providers and suppliers.

The embodiment of the invention provides a data recommendation method, which comprises the steps of obtaining online data and a recommendation rule according to the online recommendation requirement of a client; and according to the recommendation rule, screening and searching the user characteristic vector and the commodity characteristic vector in a recommendation database and an online database to generate recommendation data. The data recommendation method can provide required data for the user, reduces the screening difficulty of the user, can avoid screening errors of the user, and improves the accuracy of operation.

Example four

Fig. 4 is a flowchart illustrating a data recommendation method according to a fourth embodiment of the present invention. The embodiment is optimized based on the above embodiment, and specifically described by taking an order recommendation scheme as an example, as shown in fig. 4, the method includes:

and S410, acquiring online recommendation requirements.

The online recommendation requirement may generate requirements for online orders, for example, in shopping client software, when a user clicks on a commodity to generate a purchase order, the user often needs to fill order options, such as clothes size, color, delivery address, purchase quantity, and the like. The online order generation requirement is generated before the user clicks on the item without completing the order submission.

And S420, acquiring online data and recommendation rules from the online recommendation requirement.

At least one user characteristic in the online data may generate a user characteristic of an order user for preparation for online acquisition; at least one commodity feature in the online data is a commodity feature of a commodity to be generated, which is acquired online.

S430, searching at least one matched user characteristic vector in the database according to at least one user characteristic by adopting a recommendation rule.

Specifically, the user feature vector including the user feature may be searched, the feature vector of a specific user may be searched, or multiple feature vectors of a certain class of users may be searched.

S440, searching at least one matched commodity feature vector in the database according to at least one commodity feature.

Specifically, the method may be to find a commodity feature vector including the commodity feature, may be to find a specific commodity, or may be to find a plurality of feature vectors of a certain type of commodity.

S450, screening the searched commodity feature vector according to the searched user feature vector.

The above operation is equivalent to matching the user characteristic and the commodity characteristic, and specifically, the screening the searched commodity characteristic vector according to the searched user characteristic vector may include: according to the order type of the online order, extracting the user characteristics to be matched from the searched user characteristic vector, and extracting the commodity characteristics to be matched from the searched commodity characteristic vector; carrying out similarity matching on the characteristics of the user to be matched and the characteristics of the commodity to be matched; and screening to obtain the commodity feature vector with the similarity result reaching the set condition.

The characteristics of the user to be matched and the characteristics of the goods to be matched can be determined by the type of the order, for example, the order of clothes, the characteristics of the user to be matched can be height and weight, and the characteristics of the goods to be matched can be size. In the process of forming an order, data needs to be standardized according to national and international standards, for example: the national standard XXL corresponds to 170CM-175CM height and 60KG-70KG weight, if XXXL of a certain type of commodity is equivalent to the national standard XXL, the commodity type needs to be standardized; when some models do not meet a certain national standard or international standard model, calculating historical purchase data of the model, and finally standardizing the model into a standard model according to a probability theory. Or, according to national standards, international standards and other related data of various countries, performing similar related processing on various standardized data, and uniformly mapping the standardized data into the national standard data. For example, the data of the European code 37 of the shoe size, Chinese code 23, American code 5, British code 4.5 and the like are all 230mm in foot length, and the commodity data can be mapped into different sizes in different countries according to a data mapping system.

If the order type of the online order is electronic equipment, the characteristics of the user to be matched comprise age or preference tendency of the electronic equipment, and the characteristics of the commodity to be matched can comprise performance attributes or interest tags.

And S460, extracting recommendation data from the screened commodity feature vectors.

Extracting recommendation data from the screened commodity feature vector may include: and extracting recommendation data from the screened commodity feature vectors, filling the recommendation data into a candidate item of the online order, and displaying the candidate item as a recommended order to the user.

When a user purchases commodities, determining the characteristics of the user according to online recommendation requirements, wherein the characteristics of the user can be the height, the weight, the waist circumference, the monthly consumption amount and the type of purchased commodities, and searching a characteristic vector related to the characteristics of the user in a recommendation database or an online database according to the characteristics of the user; according to the commodity characteristics input when a user purchases commodities, searching a characteristic vector related to the commodity characteristics selected by the user in a recommendation database or an online database; and screening the characteristic vector of the commodity selected by the user according to the characteristic vector of the user, generating recommendation data and recommending the recommendation data to the user for the user to select. Meanwhile, after a user selects a certain commodity, the recommendation database performs similarity matching on the characteristics of the user to be matched and the characteristics of the commodity to be matched according to the order type of the online order without spending a large amount of time on selecting various definitional attributes of the commodity; and screening to obtain commodity characteristic vectors with similarity results reaching set conditions, extracting recommendation data from the screened commodity characteristic vectors, filling the recommendation data into the to-be-selected items of the online orders, and displaying the recommendation data serving as the recommendation orders to the user. And selecting different characteristics of the user to screen the commodities according to different commodities selected by the user.

The embodiment of the invention provides a data recommendation method which can be used for an order recommendation process, wherein user characteristics, commodity characteristics and recommendation rules in online data are obtained according to online recommendation requirements, user characteristic vectors and commodity characteristic vectors are obtained according to the recommendation rules, the searched commodity characteristic vectors are screened according to the searched user characteristic vectors, recommendation data are extracted from the screened commodity characteristic vectors, and the recommendation data are filled into a candidate of an online order and displayed to a user as a recommendation order. The data recommendation method simplifies the order placing process of the user and reduces the input operation of the user, thereby improving the order placing efficiency and the order forming rate; the problem that the user is easy to write wrong data or select wrong data when the user is not paying attention to the order is avoided, and the order return rate is reduced; and the problem that the user has no reference data and is difficult to select when ordering can be avoided.

EXAMPLE five

Fig. 5 is a schematic flowchart of a data recommendation method according to a fifth embodiment of the present invention. The present embodiment is optimized based on the above embodiments, and specifically described by taking inventory prediction as an example, as shown in fig. 5, the method includes:

and S510, acquiring online recommendation requirements.

The online recommendation demand is an inventory forecast demand; at least one commodity feature in the online data is a commodity attribute to be predicted, and the target recommendation feature is a unit time sales amount.

S520, acquiring online data and recommendation rules from the online recommendation requirements.

S530, searching at least one matched commodity feature vector in the database according to at least one commodity feature by adopting a recommendation rule.

And S540, searching at least one user characteristic vector in the database according to the searched commodity characteristic vector.

And S550, taking the target recommendation feature in the online data as a statistical dimension, and performing prediction calculation based on the commodity attention feature of the searched user feature vector to serve as recommendation data.

As will be briefly described below by taking inventory prediction as an example, as is well known, with the rapid development of online shopping, a new problem is brought, namely, the inventory storage management efficiency is low, which may cause a phenomenon of insufficient supply and demand of goods or overstock of goods, and in order to better solve the problem, the present embodiment provides a data recommendation method. Acquiring online data and a recommendation rule according to online recommendation requirements, searching commodity characteristic vectors and user characteristic vectors according to the recommendation rule, taking the sales per unit time in the online data as a statistical dimension, and performing prediction calculation based on commodity attention characteristics of the searched user characteristic vectors to serve as recommendation data; the method can also be used for analyzing according to behavior data of the user, such as data of a user browsing page, data of a user searching, data of a shopping cart article added by the user and attribute data of a commodity selected by the user, intelligently deciding and calculating the data of the user behavior analysis result and the recommendation data, calculating the purchase probability of various different attributes of the commodity according to a behavior correlation model and a behavior psychology model, finally obtaining the recommendation data of a certain commodity of a certain user, and providing the recommendation data to a supplier.

The embodiment of the invention provides a data recommendation method which can be applied to the inventory prediction process, on-line data and recommendation rules are obtained according to on-line recommendation requirements, relevant commodity feature vectors and user feature vectors are searched, target recommendation features are used as statistical dimensions, prediction calculation is carried out on the basis of commodity attention features of the searched user feature vectors, the target recommendation features are used as recommendation data, and the recommendation data are recommended to commodity suppliers. Therefore, the inventory warehousing management efficiency is improved, and the quantity of inventory commodities is reduced, so that the cost and the fund occupation are reduced; the logistics efficiency is improved, particularly the logistics management and transportation efficiency of cross-warehouse rooms is improved; the accuracy of predicting the future commodity condition is also improved.

EXAMPLE six

Fig. 6 is a schematic structural diagram of a database creating apparatus according to a sixth embodiment of the present invention. As shown in fig. 6, the apparatus includes:

a historicaldata acquisition module 610, configured to acquire user historical data and commodity historical data from a data source; the commodityfeature obtaining module 620 is configured to extract user features according to user history data and commodity history data and preset user dimensions to form a user feature vector of a user, where the user feature vector includes at least one commodity feature; the usercharacteristic obtaining module 630 is configured to extract, according to the user history data and the commodity history data, commodity characteristics according to preset commodity dimensions to form a primary commodity characteristic vector of the commodity, where the primary commodity characteristic vector includes at least one user characteristic; the commodityfeature processing module 640 is configured to combine the feature vectors of the primary commodities according to a preset commodity association rule to form a secondary commodity feature vector; and the recommendationdata forming module 650 is configured to store the user feature vector, the primary commodity feature vector, and the secondary commodity feature vector as a recommendation data set in the database.

Further, the user history data comprises user dynamic behavior data and user static attribute data; the user dynamic behavior data comprises at least one of a commodity browsing behavior, a navigation positioning behavior, a search engine searching behavior, a search result clicking behavior and a webpage browsing behavior; the user static attribute data comprises at least one of name, gender, address, height, weight and school calendar; the commodity history data includes commodity order data and commodity inventory data.

Further, the commodityfeature processing module 640 is specifically configured to: and combining the primary commodity feature vectors of which the feature value similarity of the core features in the primary commodity feature vectors reaches the set conditions according to at least one core feature set in the preset commodity association rule to form a secondary commodity feature vector.

Further, the core features include: the commodity self attribute, commodity sales data, commodity inventory data or commodity sales correlation.

Further, the apparatus further comprises: the recommendation data acquisition module is used for generating recommendation data according to online recommendation requirements and the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector in the database; and the recommendation data updating module is used for correcting the characteristic values in the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector according to the online response data aiming at the recommendation data.

Further, the apparatus further comprises: and the association rule updating module is used for receiving the updated commodity association rule input by the administrator and updating the characteristic values of the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector in the database according to the updated commodity association rule.

Further, the goods association rule includes at least one of: merging rules of the same type of commodities by taking the set commodity characteristics as classification bases; and taking the commodity characteristics with the associated behaviors as commodity association rules, wherein the associated behaviors comprise purchasing behaviors or using behaviors.

The embodiment provides a database establishing device, which obtains historical data of a user and a commodity through a historicaldata obtaining module 610, and sends the historical data to a commodityfeature obtaining module 620 and a userfeature obtaining module 630, wherein the commodityfeature obtaining module 620 is used for extracting commodity features and the userfeature obtaining module 630 is used for extracting user features, and respectively sends the extracted commodity features and the extracted user features to a commodityfeature processing module 640, and the commodityfeature processing module 640 combines feature vectors of all primary commodities according to a preset commodity association rule to form a secondary commodity feature vector; the recommendationdata forming module 650 stores the user feature vector, the primary commodity feature vector, and the secondary commodity feature vector as a recommendation data set in the database. By counting the commodity characteristics, the characteristics of people are considered in the recommendation process, and the characteristics of commodities are considered, so that the recommendation data are more accurate.

The database establishing device provided by the embodiment of the invention and the database establishing method provided by any embodiment of the invention belong to the same inventive concept, can execute the database establishing method provided by any embodiment of the invention, and have corresponding functions and beneficial effects. For technical details that are not described in detail in this embodiment, reference may be made to a database establishment method provided in any embodiment of the present invention.

EXAMPLE seven

Fig. 7 is a schematic structural diagram of a data recommendation device according to a seventh embodiment of the present invention. As shown in fig. 7, the apparatus includes:

arequirement obtaining module 710, configured to obtain an online recommendation requirement; aninformation obtaining module 720, configured to obtain online data and recommendation rules from online recommendation requirements; theinformation processing module 730 is configured to generate recommendation data based on the online data, the user feature vector in the database, and the commodity feature vector by using a recommendation rule, where the user feature vector includes at least one commodity feature, and the commodity feature vector includes at least one user feature.

Further, theinformation processing module 730 includes: the online data acquisition module is used for acquiring online data from the online recommendation requirement, wherein the online data comprises at least one user characteristic and/or at least one commodity characteristic; and the recommendation rule acquisition module is used for searching a corresponding recommendation rule according to the online recommendation requirement.

Further, theinformation processing module 730 includes: the user characteristic vector acquisition module is used for searching at least one matched user characteristic vector in the database according to at least one user characteristic by adopting a recommendation rule; the commodity feature vector acquisition module is used for searching at least one matched commodity feature vector in the database according to at least one commodity feature; the screening module is used for screening the searched commodity feature vectors according to the searched user feature vectors; and the data recommendation module is used for extracting recommendation data from the screened commodity feature vectors.

Further, the online recommendation demand generates a demand for the online order; at least one user characteristic in the online data is a user characteristic of a user who prepares to generate an order for online acquisition; at least one commodity feature in the online data is a commodity feature of a commodity to be generated, which is acquired online; correspondingly, the screening module is specifically configured to: according to the order type of the online order, extracting the user characteristics to be matched from the searched user characteristic vector, and extracting the commodity characteristics to be matched from the searched commodity characteristic vector; carrying out similarity matching on the user characteristics to be matched and the commodity characteristics to be matched; and screening to obtain the commodity feature vector with the similarity result reaching the set condition.

Further, if the order type of the online order is clothes, the characteristics of the user to be matched comprise height or weight, and the characteristics of the commodity to be matched comprise size; or the order type of the online order is electronic equipment, the characteristics of the user to be matched comprise age or preference tendency of the electronic equipment, and the characteristics of the commodity to be matched comprise performance attributes or interest labels.

Further, the data recommendation module is further configured to: and extracting recommendation data from the screened commodity feature vectors, filling the recommendation data into a candidate item of the online order, and displaying the candidate item as a recommendation order to the user.

Further, theinformation processing module 730 includes: the commodity feature vector acquisition module is used for searching at least one matched commodity feature vector in the database according to at least one commodity feature by adopting a recommendation rule; the user characteristic vector acquisition module is used for searching at least one user characteristic vector in the database according to the searched commodity characteristic vector; and the data recommendation module is used for taking the target recommendation characteristics in the online data as statistical dimensions, and performing prediction calculation based on the commodity attention characteristics of the searched user characteristic vector to serve as recommendation data.

Further, the online recommendation demand is an inventory forecast demand; at least one commodity feature in the online data is a commodity attribute to be predicted, and the target recommendation feature is a unit time sales amount.

According to the data recommendation device provided by the embodiment of the invention, the online recommendation requirement is acquired through therequirement acquisition module 710 and is sent to theinformation acquisition module 720, theinformation acquisition module 720 acquires online data and a recommendation rule according to the recommendation requirement, and theinformation processing module 730 generates recommendation data by adopting the recommendation rule based on the online data, the user characteristic vector and the commodity characteristic vector in the database. The data recommendation device can provide required data for the user, reduces the screening difficulty of the user, can avoid screening errors of the user, and improves the accuracy of operation.

The data recommendation device provided by the embodiment belongs to the same inventive concept as the data recommendation method provided by any embodiment of the invention, can execute the data recommendation method provided by any embodiment of the invention, and has corresponding functions and beneficial effects. For technical details that are not described in detail in this embodiment, reference may be made to a data recommendation method provided in any embodiment of the present invention.

An embodiment of the present invention further provides an apparatus, where the apparatus includes: one or more processors; a storage device, configured to store one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the database establishment method provided in any embodiment of the present invention.

An embodiment of the present invention provides another apparatus, where the apparatus includes: one or more processors;

a storage device, configured to store one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the database recommendation method provided in any embodiment of the present invention.

The structure of the above apparatus is explained by the following embodiment eight.

Example eight

Fig. 8 is a schematic structural diagram of an apparatus according to an eighth embodiment of the present invention. Fig. 8 illustrates a block diagram of anexemplary device 8100, which is suitable for use in implementing embodiments of the present invention. Thedevice 8100 shown in fig. 8 is only an example and should not bring any limitations to the function and the scope of use of the embodiments of the present invention.

As shown in fig. 8,device 8100 is in the form of a general purpose computing device. Components ofdevice 8100 may include, but are not limited to: one or more processors orprocessing units 8110, asystem memory 8120, and abus 8130 that couples the various system components including thesystem memory 8120 and theprocessing unit 8110.

Bus 8130 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Device 8100 typically includes a variety of computer system readable media. Such media can be any available media that is accessible bydevice 8100 and includes both volatile and nonvolatile media, removable and non-removable media.

Thesystem memory 8120 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)8121 and/orcache memory 8122.Device 8100 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only,storage system 8123 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 8, and commonly referred to as a "hard drive"). Although not shown in FIG. 8, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In such cases, each drive may be connected tobus 8130 by one or more data media interfaces.Memory 8120 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.

A program/utility 8125 having a set (at least one) ofprogram modules 8124 may be stored, for example, inmemory 8120,such program modules 8124 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may include an implementation of a network environment.Program modules 8124 generally perform the functions and/or methodologies of the described embodiments of the invention.

Thedevice 8100 may also communicate with one or more external devices 8300 (e.g., keyboard, pointing device,display 8200, etc.), with one or more devices that enable a user to interact with thedevice 8100, and/or with any devices (e.g., network card, modem, etc.) that enable thedevice 8100 to communicate with one or more other computing devices. Such communication may occur via an input/output (I/O)interface 8140. Also,device 8100 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the internet) vianetwork adapter 8150. As shown, thenetwork adapter 8150 communicates with the other modules of thedevice 8100 via abus 8130. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with thedevice 8100, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

Theprocessing unit 8110 executes programs stored in thesystem memory 8120, thereby executing various functional applications and data processing, for example, implementing a database establishing method or a data recommending method provided by an embodiment of the present invention.

Embodiments of the present invention provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform a database building method provided by any of the embodiments of the present invention.

Yet another storage medium containing computer-executable instructions for performing a database recommendation method provided by any of the embodiments of the present invention when executed by a computer processor is provided.

Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims (20)

1. A method for establishing a database, comprising:

acquiring user history data and commodity history data from a data source;

extracting user features according to user historical data and commodity historical data and preset user dimensions to form a user feature vector of a user, wherein the user feature vector comprises at least one commodity feature;

according to the user historical data and the commodity historical data, commodity features are extracted according to preset commodity dimensions to form a primary commodity feature vector of the commodity, wherein the primary commodity feature vector comprises at least one user feature;

combining the characteristic vectors of the primary commodities according to a preset commodity association rule to form a characteristic vector of a secondary commodity;

storing the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector as a recommendation data set in a database;

generating recommendation data according to online recommendation requirements and the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector in the database;

and receiving an updated commodity association rule input by an administrator, and updating the characteristic values of the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector in the database based on the updated commodity association rule according to the online response data aiming at the recommendation data.

2. The method of claim 1, wherein:

the user historical data comprises user dynamic behavior data and user static attribute data;

the user dynamic behavior data comprises at least one of a commodity browsing behavior, a navigation positioning behavior, a search engine searching behavior, a search result clicking behavior and a webpage browsing behavior; the user static attribute data comprises at least one of name, gender, address, height, weight and school calendar;

the commodity history data includes commodity order data and commodity inventory data.

3. The method of claim 1, wherein merging the primary commodity feature vectors to form secondary commodity feature vectors according to a preset commodity association rule comprises:

and combining the primary commodity feature vectors of which the feature value similarity of the core features in the primary commodity feature vectors reaches the set conditions according to at least one core feature set in the preset commodity association rule to form a secondary commodity feature vector.

4. The method of claim 3, wherein the core features comprise: the commodity self attribute, commodity sales data, commodity inventory data or commodity sales correlation.

5. The method of any of claims 1-4, wherein the merchandise association rules include at least one of:

merging rules of the same type of commodities by taking the set commodity characteristics as classification bases;

and taking the commodity characteristics with the associated behaviors as commodity association rules, wherein the associated behaviors comprise purchasing behaviors or using behaviors.

6. A method for recommending data, comprising:

acquiring an online recommendation requirement; the online recommendation requirement is an online order generation requirement;

acquiring online data and recommendation rules from the online recommendation requirements; wherein the online data comprises at least one user characteristic and/or at least one merchandise characteristic; at least one user characteristic in the online data is a user characteristic of a user who prepares to generate an order for online acquisition; at least one commodity feature in the online data is a commodity feature of a commodity to be generated, which is acquired online;

searching at least one matched user characteristic vector in a database according to at least one user characteristic by adopting the recommendation rule; wherein the user feature vector comprises at least one commodity feature;

searching at least one matched commodity feature vector in a database according to the at least one commodity feature; wherein the commodity feature vector comprises at least one user feature;

according to the order type of the online order, extracting the user characteristics to be matched from the searched user characteristic vector, and extracting the commodity characteristics to be matched from the searched commodity characteristic vector;

carrying out similarity matching on the user characteristics to be matched and the commodity characteristics to be matched;

screening to obtain commodity feature vectors with similarity results meeting set conditions;

and extracting recommendation data from the screened commodity feature vectors.

7. The method of claim 6, wherein obtaining online data and recommendation rules from the online recommendation need comprises:

acquiring online data from the online recommendation requirement;

and searching a corresponding recommendation rule according to the online recommendation requirement.

8. The method of claim 6, wherein:

if the order type of the online order is clothes, the characteristics of the user to be matched comprise height or weight, and the characteristics of the commodity to be matched comprise size; or

If the order type of the online order is electronic equipment, the characteristics of the user to be matched comprise age or preference tendency of the electronic equipment, and the characteristics of the commodity to be matched comprise performance attributes or interest tags.

9. The method of claim 6, wherein extracting recommendation data from the filtered commodity feature vector comprises:

and extracting recommendation data from the screened commodity feature vectors, filling the recommendation data into a candidate item of the online order, and displaying the candidate item as a recommendation order to the user.

10. The method of claim 7, wherein generating recommendation data based on the online data, the user feature vector in the database, and the commodity feature vector using the recommendation rule comprises:

searching at least one matched commodity feature vector in a database according to at least one commodity feature by adopting the recommendation rule;

searching at least one user characteristic vector in a database according to the searched commodity characteristic vector;

and taking the target recommendation characteristics in the online data as statistical dimensions, and performing prediction calculation based on the commodity attention characteristics of the searched user characteristic vector to serve as recommendation data.

11. The method of claim 10, wherein:

the online recommendation demand is an inventory forecast demand;

at least one commodity feature in the online data is a commodity attribute to be predicted, and the target recommendation feature is a unit time sales amount.

12. A database building apparatus, comprising:

the historical data acquisition module is used for acquiring user historical data and commodity historical data from a data source;

the commodity feature acquisition module is used for extracting user features according to user historical data and commodity historical data and preset user dimensions to form a user feature vector of a user, wherein the user feature vector comprises at least one commodity feature;

the system comprises a user characteristic acquisition module, a commodity dimension extraction module and a commodity dimension extraction module, wherein the user characteristic acquisition module is used for extracting commodity characteristics according to user historical data and commodity historical data and preset commodity dimensions to form a primary commodity characteristic vector of a commodity, and the primary commodity characteristic vector comprises at least one user characteristic;

the commodity feature processing module is used for merging the characteristic vectors of the primary commodities according to a preset commodity association rule to form a characteristic vector of a secondary commodity;

the recommendation data forming module is used for storing the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector as a recommendation data set in a database;

the recommendation data acquisition module is used for generating recommendation data according to online recommendation requirements and the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector in the database;

the recommendation data updating module is used for correcting the characteristic values in the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector according to the online response data aiming at the recommendation data;

and the association rule updating module is used for receiving the updated commodity association rule input by the administrator and updating the characteristic values of the user characteristic vector, the primary commodity characteristic vector and the secondary commodity characteristic vector in the database according to the updated commodity association rule.

13. The apparatus according to claim 12, wherein the product characteristic processing module is specifically configured to:

and combining the primary commodity feature vectors of which the feature value similarity of the core features in the primary commodity feature vectors reaches the set conditions according to at least one core feature set in the preset commodity association rule to form a secondary commodity feature vector.

14. A data recommendation device, comprising:

the demand acquisition module is used for acquiring online recommendation demands; the online recommendation requirement is an online order generation requirement;

the information acquisition module is used for acquiring online data and recommendation rules from the online recommendation requirements; wherein the online data comprises at least one user characteristic and/or at least one merchandise characteristic; at least one user characteristic in the online data is a user characteristic of a user who prepares to generate an order for online acquisition; at least one commodity feature in the online data is a commodity feature of a commodity to be generated, which is acquired online;

the information processing module is used for generating recommendation data based on online data, a user feature vector and a commodity feature vector in a database by adopting the recommendation rule, wherein the user feature vector comprises at least one commodity feature, and the commodity feature vector comprises at least one user feature;

wherein, the information processing module comprises:

the user characteristic vector acquisition module is used for searching at least one matched user characteristic vector in a database according to at least one user characteristic by adopting the recommendation rule;

the commodity feature vector acquisition module is used for searching at least one matched commodity feature vector in a database according to the at least one commodity feature;

the screening module is used for screening the searched commodity characteristic vector according to the searched user characteristic vector;

the data recommendation module is used for extracting recommendation data from the screened commodity feature vectors;

wherein, the screening module is specifically configured to: according to the order type of the online order, extracting the user characteristics to be matched from the searched user characteristic vector, and extracting the commodity characteristics to be matched from the searched commodity characteristic vector; carrying out similarity matching on the characteristics of the user to be matched and the characteristics of the commodity to be matched; and screening to obtain the commodity feature vector with the similarity result reaching the set condition.

15. The apparatus of claim 14, wherein the information processing module comprises:

the online data acquisition module is used for acquiring online data from the online recommendation requirement;

and the recommendation rule acquisition module is used for searching a corresponding recommendation rule according to the online recommendation requirement.

16. The apparatus of claim 15, wherein the information processing module comprises:

the commodity feature vector acquisition module is used for searching at least one matched commodity feature vector in a database according to at least one commodity feature by adopting the recommendation rule;

the user characteristic vector acquisition module is used for searching at least one user characteristic vector in the database according to the searched commodity characteristic vector;

and the data recommendation module is used for taking the target recommendation characteristics in the online data as statistical dimensions, and performing prediction calculation based on the commodity attention characteristics of the searched user characteristic vector to serve as recommendation data.

17. A database building apparatus, characterized in that the database building apparatus comprises:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the database building method of any one of claims 1-5.

18. A data recommendation device, characterized in that the data recommendation device comprises:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement a data recommendation method as claimed in any one of claims 6-11.

19. A storage medium containing computer-executable instructions for performing the database building method of any one of claims 1-5 when executed by a computer processor.

20. A storage medium containing computer-executable instructions for performing the data recommendation method of any one of claims 6-11 when executed by a computer processor.

Images