Summary of the invention
In view of the above problems, it proposes the embodiment of the present invention and overcomes the above problem or at least partly in order to provide one kindA kind of recommended method of the cell dictionary to solve the above problems and a kind of corresponding recommendation apparatus of cell dictionary.
To solve the above-mentioned problems, the embodiment of the invention discloses a kind of recommended methods of cell dictionary, comprising:
Determine that mounted cell dictionary, the mounted cell dictionary have corresponding feature score value；
Obtain at least one corresponding cell dictionary to be recommended of each mounted cell dictionary, the cell word to be recommendedLibrary has corresponding similarity with described installed between cell dictionary；
Using the feature score value of the mounted cell dictionary and the similarity of the cell dictionary to be recommended, calculate everyThe linked character score value of a cell dictionary to be recommended；
According to the linked character score value, target cell dictionary is extracted from the cell dictionary to be recommended；
Recommend the target cell dictionary.
Optionally, described the step of obtaining each mounted cell dictionary corresponding at least one cell dictionary to be recommendedInclude:
Calculate the similarity between each mounted cell dictionary and other cell dictionaries；
At least one the cell dictionary of the similarity more than preset threshold is extracted as cell dictionary to be recommended.
Optionally, the step of similarity calculated between each mounted cell dictionary and other cell dictionaries is wrappedIt includes:
Obtain the user characteristics score value of each cell dictionary；
According to the user characteristics score value of each cell dictionary, the feature vector of each cell dictionary is generated；
Using described eigenvector, the similarity between each mounted cell dictionary and other cell dictionaries is calculated.
Optionally, the step of user characteristics score value for obtaining each cell dictionary includes:
Set the initial characteristics score value of each mounted cell dictionary；
It is corresponding to the mounted cell dictionary when the word shielded on user is the word in mounted cell dictionaryInitial characteristics score value is incremented by；
It is corresponding to the mounted cell dictionary when the word that user deletes is the word in mounted cell dictionaryInitial characteristics score value successively decreases；
User characteristics score value with revised initial characteristics score value, as the mounted cell dictionary.
Optionally, described to use described eigenvector, calculate each mounted cell dictionary and other cell dictionaries itBetween similarity the step of include:
It calculates separately between the feature vector of each mounted cell dictionary and the feature vector of other cell dictionariesDistance, using the distance as the similarity between each mounted cell dictionary and other cell dictionaries.
Optionally, described using the feature score value of the mounted cell dictionary and the phase of the cell dictionary to be recommendedLike degree, the step of calculating the linked character score value of each cell dictionary to be recommended, includes:
Using the similarity product of the feature score value of the mounted cell dictionary and the cell dictionary to be recommended asThe initial association feature score value of the cell dictionary to be recommended；
The corresponding whole initial association feature score values of each cell dictionary to be recommended are added up, are obtained described each to be recommended thinThe linked character score value of born of the same parents' dictionary.
Optionally, described according to the linked character score value, target cell word is extracted from the cell dictionary to be recommendedThe step of library includes:
The cell dictionary to be recommended is ranked up according to the linked character score value；
The cell dictionary to be recommended for extracting the preceding preset quantity that sorts is target cell dictionary.
To solve the above-mentioned problems, the embodiment of the invention discloses a kind of recommendation apparatus of cell dictionary, comprising:
Determining module, for determining that mounted cell dictionary, the mounted cell dictionary have corresponding featureScore value；
Module is obtained, for obtaining at least one corresponding cell dictionary to be recommended of each mounted cell dictionary, instituteStating cell dictionary to be recommended has corresponding similarity with described installed between cell dictionary；
Computing module, for the feature score value and the cell dictionary to be recommended using the mounted cell dictionarySimilarity calculates the linked character score value of each cell dictionary to be recommended；
Extraction module, for extracting target cell from the cell dictionary to be recommended according to the linked character score valueDictionary；
Recommending module, for recommending the target cell dictionary.
Optionally, the acquisition module includes:
Similarity calculation submodule, it is similar between each mounted cell dictionary and other cell dictionaries for calculatingDegree；
Cell dictionary extracting sub-module to be recommended, for extracting at least one cell that the similarity is more than preset thresholdDictionary is as cell dictionary to be recommended.
Optionally, the similarity calculation submodule includes:
User characteristics score value acquiring unit, for obtaining the user characteristics score value of each cell dictionary；
Feature vector generation unit generates each cell for the user characteristics score value according to each cell dictionaryThe feature vector of dictionary；
Similarity calculated calculates each mounted cell dictionary and other is thin for using described eigenvectorSimilarity between born of the same parents' dictionary.
Optionally, the user characteristics score value acquiring unit includes:
First setting subelement, for setting the initial characteristics score value of each mounted cell dictionary；
It is incremented by subelement, for having been installed to described when the word shielded on user is the word in mounted cell dictionaryThe corresponding initial characteristics score value of cell dictionary be incremented by；
Successively decrease subelement, for having been installed to described when the word that user deletes is the word in mounted cell dictionaryThe corresponding initial characteristics score value of cell dictionary successively decrease；
User characteristics score value determines subelement, is used for revised initial characteristics score value, as described mounted thinThe user characteristics score value of born of the same parents' dictionary.
Optionally, the similarity calculated includes:
Similarity calculation subelement, for calculating separately the feature vector and other cells of each mounted cell dictionaryThe distance between feature vector of dictionary, using the distance as between each mounted cell dictionary and other cell dictionariesSimilarity.
Optionally, the computing module includes:
Initial association feature score value computational submodule, for the feature score value of the mounted cell dictionary and describedInitial association feature score value of the similarity product of cell dictionary to be recommended as the cell dictionary to be recommended；
Linked character score value computational submodule, it is special for adding up the corresponding whole initial associations of each cell dictionary to be recommendedScore value is levied, the linked character score value of each cell dictionary to be recommended is obtained.
Optionally, the extraction module includes:
Sorting sub-module, for being ranked up according to the linked character score value to the cell dictionary to be recommended；
Extracting sub-module, the cell dictionary to be recommended for extracting the preceding preset quantity that sorts are target cell dictionary.
To solve the above-mentioned problems, the embodiment of the invention discloses a kind of recommendation apparatus of cell dictionary, include storagePerhaps more than one program one of them or more than one program is stored in memory by device and one, and is configuredIt include for performing the following operation to execute the one or more programs by one or more than one processorInstruction:
Determine that mounted cell dictionary, the mounted cell dictionary have corresponding feature score value；
Obtain at least one corresponding cell dictionary to be recommended of each mounted cell dictionary, the cell word to be recommendedLibrary has corresponding similarity with described installed between cell dictionary；
Using the feature score value of the mounted cell dictionary and the similarity of the cell dictionary to be recommended, calculate everyThe linked character score value of a cell dictionary to be recommended；
According to the linked character score value, target cell dictionary is extracted from the cell dictionary to be recommended；
Recommend the target cell dictionary.
To solve the above-mentioned problems, the embodiment of the invention discloses a kind of storage mediums, the finger in the storage mediumWhen enabling the processor execution by terminal, the recommendation side for executing said one or multiple cell dictionaries is enabled the terminal toMethod.
Compared with the background art, the embodiment of the present invention includes following advantages:
The embodiment of the present invention, it is each mounted thin by obtaining after determining the mounted cell dictionary of active userThen the corresponding cell dictionary to be recommended of born of the same parents' dictionary uses the feature score value and cell dictionary to be recommended of mounted cell dictionarySimilarity, the linked character score value of each cell dictionary to be recommended is calculated, so as to according to linked character score value, from wait push awayIt recommends and extracts target cell dictionary in cell dictionary, and target cell dictionary is recommended into user, solve in the prior art onlyThe problem of capable of just recommending the cell dictionary after the vocabulary that user inputs in repeatedly some cell dictionary.The embodiment of the present inventionIt can be by determining user's currently mounted cell dictionary, extracting has higher similarity with mounted cell dictionaryCell dictionary, and user is recommended, just it can recommend cell to user before the input of user, it is convenient for users to use, it mentionsHigh user's input efficiency.Meanwhile the embodiment of the present invention can recommend more relevant under the premise of not reducing accuracyCell dictionary also improves the utilization rate of cell dictionary.
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific realApplying mode, the present invention is described in further detail.
Referring to Fig.1, a kind of step flow chart of the recommended method embodiment one of cell dictionary of the invention is shown, specificallyIt may include steps of:
Step 101, determine that mounted cell dictionary, the mounted cell dictionary have corresponding feature score value；
In the concrete realization, the embodiment of the present invention can be applied in each Terminal Type, for example, mobile phone, PDA (PersonalDigital Assistant, personal digital assistant), computer, palm PC etc., the embodiment of the present invention is to the specific of terminalType is not construed as limiting.These terminals can support to include Windows, Android (Android), IOS, WindowsPhone etc. a variety ofThe operating system of type.
In general, in order to meet the personalized input demand of user, input method provides the function of cell dictionary, and user can be withBy installation specific area or the cell dictionary of particular category, simplify input process.For example, for doctor, due to being frequently necessary toThe relevant vocabulary of medical domain is inputted, can choose the cell dictionary of installation medicine；For basketball fan, if be frequently necessary to defeatedEnter with NBA (National Basketball Association, the league matches of U.S. man professional basketball) relevant vocabulary, can be withConsider the cell dictionary of installation NBA.
Therefore, in embodiments of the present invention, it can determine that mounted cell dictionary has active user at the terminal firstWhich.Each mounted cell dictionary can have corresponding feature score value, and this feature score value can embody user's useThe frequency of the cell dictionary.For example, an initial value can be set for each mounted cell dictionary, when the word life shielded on userIn some mounted cell dictionary when, can to the initial value be incremented by；Conversely, having been installed when the word that user deletes hits someCell dictionary when, can successively decrease to the initial value, to obtain the feature score value of the cell dictionary.
It should be noted that since service condition of each user to each cell dictionary differs greatly, Mei GexiThe feature score value of born of the same parents' dictionary is different according to the difference of user.For example, for the same cell dictionary, for three different useFamily, feature score value may be different.
Step 102, at least one corresponding cell dictionary to be recommended of each mounted cell dictionary is obtained, it is described wait push awayRecommending cell dictionary has corresponding similarity with described installed between cell dictionary；
Typically for a certain professional domain, which can further be refined.For example, for basketball field, it canTo be further divided into NBA and CBA (China Basketball Association, the professional tournament of Men's basketball), fromAnd cell dictionary relevant to NBA and cell dictionary relevant with CBA can be formed, above-mentioned two cell dictionary is with higherSimilarity.
Therefore, in embodiments of the present invention, active user at the terminal after mounted cell dictionary is being determined, it can be withObtain the corresponding cell dictionary to be recommended of each mounted cell dictionary, the cell dictionary to be recommended can be with it is mountedCell dictionary has the cell dictionary of higher similarity.For example, NBA cell dictionary mounted for user, it is available withThe NBA cell dictionary has the CBA cell dictionary or other cell dictionaries of higher similarity.The cell dictionary to be recommended obtained canBe one, it is two or more, the embodiment of the present invention is not construed as limiting the specific number of cell dictionary to be recommended.
In the concrete realization, each mounted cell can be calculated separately according to the feature score value of each cell dictionarySimilarity between dictionary and other cell dictionaries, so that at least one the cell dictionary for extracting similarity more than preset threshold is madeFor cell dictionary to be recommended.The size of preset threshold can specifically be set by those skilled in the art, and the embodiment of the present invention is to thisIt is not construed as limiting.
Alternatively, after calculating the similarity between each mounted cell dictionary and other cell dictionaries, according to phaseLike degree descending arrangement, to extract at least one maximum cell dictionary of similarity as cell dictionary to be recommended, the present invention is realIt applies example and this is also not construed as limiting.
Step 103, using the similar of the feature score value of the mounted cell dictionary and the cell dictionary to be recommendedDegree calculates the linked character score value of each cell dictionary to be recommended；
It in the concrete realization, can be by the feature score value of mounted cell dictionary and corresponding cell dictionary to be recommendedSimilarity be multiplied, using the product of the two as the linked character score value of each cell dictionary to be recommended.
For example, the feature score value for corresponding to active user is SS1 for mounted cell dictionary S1, meanwhile, have with S1The cell dictionary to be recommended for having higher similarity includes J1, J2 and J3, the similarity difference of cell dictionary J1, J2 and J3 to be recommendedFor simJ1, simJ2 and simJ3, then the feature score value that can calculate separately out S1 is the similarity of SS1 and J1, J2 and J3Corresponding product between simJ1, simJ2 and simJ3, using SS1*simJ1 as the linked character of cell dictionary J1 to be recommended pointValue, using SS1*simJ2 as the linked character score value of cell dictionary J2 to be recommended, using SS1*simJ3 as cell word to be recommendedThe linked character score value of library J3.
It should be noted that since active user may be equipped with multiple cell dictionaries, it may for example comprise mounted cellDictionary S2 and S3, then the corresponding cell dictionary to be recommended of S1, S2 and S3 may include identical cell dictionary.For example,The corresponding cell dictionary to be recommended of S1 be J1, J2 and J3 when, if the corresponding cell dictionary to be recommended of S2 be J1, J4 and J5, S3 pairsThe cell dictionary to be recommended answered is J2, J6 and J7, then there are identical cell dictionary J1, S1 and S3 to be recommended, there are phases with S2 by S1Same cell dictionary J2 to be recommended.At this point, for duplicate removal, it can be in the association for calculating separately out each cell dictionary to be recommendedAfter feature score value, the linked character score value of identical cell dictionary to be recommended is added, as final linked character score value.
For example, for feature score value J1 to be recommended, the linked character score value relative to mounted cell dictionary S1 isSS1*simJ1, the linked character score value relative to mounted cell dictionary S2 are that (wherein, SS2 is S2 corresponding to SS2*simJ1The feature score value of active user), then linked character score value that can be final as J1 using SS1*simJ1+SS2*simJ1.
Certainly, those skilled in the art can also use other calculations, and the association for calculating cell dictionary to be recommended is specialScore value is levied, the embodiment of the present invention is not construed as limiting this.
Step 104, according to the linked character score value, target cell dictionary is extracted from the cell dictionary to be recommended；
In the concrete realization, after the linked character score value for obtaining each cell dictionary to be recommended, it is special that association can be extractedSeveral highest cell dictionaries of score value are levied as target cell dictionary.For example, can be arranged according to linked character score value descending,Several preceding cell dictionaries of selected and sorted are as target cell dictionary.The quantity of target cell dictionary can be by this field skillDetermine according to actual needs, the embodiment of the present invention is not construed as limiting this to art personnel.
Step 105, recommend the target cell dictionary.
In embodiments of the present invention, after determining target cell dictionary, initiatively the target cell dictionary can be recommendedTo user.
In the concrete realization, prompt information can be sent to user, reminds user installation target cell dictionary.For example, canTo send following information to user: " according to your service condition, you being recommended to install following cell dictionary." when user it is selected whereinWhen some recommended target cell dictionary, which can be installed, facilitate the subsequent use of user.OrPerson can also be mounted directly the target cell dictionary, the embodiment of the present invention when recommended target cell dictionary negligible amountsThe concrete mode of recommendation is not construed as limiting.
In embodiments of the present invention, it after determining the mounted cell dictionary of active user, has each been installed by obtainingThe corresponding cell dictionary to be recommended of cell dictionary, then use mounted cell dictionary feature score value and cell to be recommendedThe similarity of dictionary calculates the linked character score value of each cell dictionary to be recommended, so as to according to linked character score value, fromTarget cell dictionary is extracted in cell dictionary to be recommended, and target cell dictionary is recommended into user, solves the prior artIn the problem of can only just recommending the cell dictionary after the vocabulary that user inputs repeatedly in some cell dictionary.The present invention is realApplying example can be extracted to mounted cell dictionary by determining user's currently mounted cell dictionary with higher similarThe cell dictionary of degree, and user is recommended, it just can recommend cell to user before the input of user, facilitate making for userWith improving user's input efficiency.Meanwhile the embodiment of the present invention can recommend more under the premise of not reducing accuracyRelevant cell dictionary, also improves the utilization rate of cell dictionary.
Referring to Fig. 2, a kind of step flow chart of the recommended method embodiment two of cell dictionary of the invention is shown, specificallyIt may include steps of:
Step 201, determine that mounted cell dictionary, the mounted cell dictionary have corresponding feature score value；
In general, when user uses input method at the terminal, specific area or particular category can be installed in input methodCell dictionary simplifies input process.The terminal can be the electronic equipments such as mobile phone, tablet computer, desktop computer, and the present invention is implementedExample is not construed as limiting this.
In embodiments of the present invention, can determine which mounted cell dictionary has to active user at the terminal first.Each mounted cell dictionary can have corresponding feature score value, and this feature score value can embody user and use the cellThe frequency of dictionary.For example, if the word shielded on user more hits some mounted cell dictionary, the spy of the cell dictionaryIt is higher to levy score value.
Step 202, the similarity between each mounted cell dictionary and other cell dictionaries is calculated；
In embodiments of the present invention, in order to obtain with each mounted cell dictionary similarity with higher otherCell dictionary can obtain the user characteristics score value of each cell dictionary first, which can be according to eachUser scores to the service condition of each cell dictionary and obtains.
It should be noted that since service condition of each user to each cell dictionary differs greatly, Mei GexiThe feature score value of born of the same parents' dictionary is different according to the difference of user.For example, for the same cell dictionary, for three different useFamily, user characteristics score value may be different.
In embodiments of the present invention, the whole cell dictionaries being capable of providing for input method, can be divided into active userMounted cell dictionary and uninstalled cell dictionary.Therefore, the first of each mounted cell dictionary can be set firstBeginning feature score value.It, can be to the mounted cell dictionary when the word shielded on user is the word in mounted cell dictionaryCorresponding initial characteristics score value is incremented by；When the word that user deletes is the word in mounted cell dictionary, this can be pacifiedThe corresponding initial characteristics score value of the cell dictionary of dress successively decreases, mounted as this then with revised initial characteristics score valueThe user characteristics score value of cell dictionary.On the other hand, for uninstalled cell dictionary, uninstalled cell dictionary can be setFeature score value be a fixed value.Wherein, which can be less than the initial characteristics score value of mounted cell dictionary.
For example, the initial characteristics score value of mounted cell dictionary can be set as 1, then each mounted cell dictionaryInitial feature score value is 1, when the word shielded on user hits the cell dictionary, can add 1 to feature score value, when user deletesWhen the word removed hits the cell dictionary, can subtract 1 to feature score value, then numerical result final using in certain period of time asThe user characteristics score value of the cell dictionary.Cell dictionary uninstalled for active user, can set its feature score value isFixed value 0.Certainly, those skilled in the art can also set the feature score value of each cell dictionary, this hair using other modesBright embodiment is not construed as limiting this.
It, can use by each user of client simulation to each cell dictionary it should be noted that in practiceSituation, to obtain the user characteristics score value of each cell dictionary.
In the concrete realization, input method client can with the user characteristics score value of each cell dictionary of periodic statistical, for example,The user characteristics score value of primary each cell dictionary can be counted with every 24 hours, and the user characteristics score value is sent to serviceDevice.Then, server can generate the feature vector of each cell dictionary according to the user characteristics score value of each cell dictionary,And this feature vector is used, calculate the similarity between each mounted cell dictionary and other cell dictionaries.Specifically, may be usedTo calculate separately the feature vector of each mounted cell dictionary and the distance between the feature vector of other cell dictionaries, withThe distance is as the similarity between each mounted cell dictionary and other cell dictionaries.
This feature can also be used after the feature vector for generating each cell dictionary as a kind of example of the inventionVector calculates separately the similarity between each cell dictionary and other cell dictionaries, then according to similarity, cellulation wordLibrary similarity list.Thus when needing to calculate the similarity between mounted cell dictionary and other cell dictionaries, directlyFrom the similarity extracted in the similarity list between mounted cell dictionary and other cell dictionaries.
It specifically, can be the quantity of all users in the big matrix of one N*M of server end maintenance, N, M is all thinThe sum of born of the same parents' dictionary, matrix can be expressed as follows:
Wherein, scel1, scel2 ... scelM indicate that M cell dictionary, user1, user2 ... userN indicate N number ofUser, S11 indicate cell dictionary scel1 relative to the feature score value of user user1, S12 indicate cell dictionary scel2 relative toFeature score value, the SNM of user user1 indicates feature score value of the cell dictionary scelM relative to user userN, and so on.
In the matrix, every a line indicates that a cell dictionary corresponds to the user characteristics score value of all users, Mei YilieThen indicate that all cell dictionaries correspond to the user characteristics score value of a user.So as to calculate the phase between cell dictionaryWhen seemingly spending, using the feature score value of every a line as a vector, N number of feature of the cell dictionary is indicated, thus by calculating twoThe distance between a vector calculates the similarity obtained between two cell dictionaries.Calculate two vectors between apart from when,Can using Pearson correlation coefficient (Pearson product-moment correlation coefficient) or otherThe calculation of related coefficient, the embodiment of the present invention are not construed as limiting this.
It should be noted that calculate cell dictionary between similarity frequency do not need too frequently, can be with every 5 daysPeriod calculates once, and calculated result is saved in the server.Certainly, those skilled in the art can according to actual needs,Specific to determine the period for uploading feature score value and calculating similarity, the embodiment of the present invention is not construed as limiting this.
Step 203, at least one the cell dictionary of the similarity more than preset threshold is extracted as cell word to be recommendedLibrary；
In embodiments of the present invention, it is calculating separately out between each mounted cell dictionary and other cell dictionariesIt can be more than the cell dictionary of preset threshold using similarity as cell dictionary to be recommended after similarity.For example, can set pre-It, can be with when the similarity between some cell dictionary and some mounted cell dictionary is more than 75% if threshold value is 75%Using the cell dictionary as the cell dictionary to be recommended of the mounted cell dictionary.
It, should be to it should be noted that each mounted cell dictionary can have at least one cell word to be recommendedThe cell dictionary of recommendation not only can be the uninstalled cell dictionary of active user, can also be with the mounted cell word of active userLibrary.On the other hand, the corresponding cell to be recommended of the mounted cell dictionary of any two is also possible to identical, and active user isThe corresponding cell dictionary to be recommended of the cell dictionary S1 of installation is J1, J2 and J3, and mounted cell dictionary S2 is corresponding wait push awayRecommending cell dictionary is J1, J4 and J5, and the two all has identical cell dictionary J1 to be recommended.
Step 204, with the similarity of the feature score value of the mounted cell dictionary and the cell dictionary to be recommendedInitial association feature score value of the product as the cell dictionary to be recommended；
It in embodiments of the present invention, can be by the feature score value of mounted cell dictionary and corresponding cell to be recommendedThe similarity of dictionary is multiplied, using the product of the two as the initial association feature score value of each cell dictionary to be recommended.
In the concrete realization, the initial characteristics score value of each cell dictionary to be recommended can be calculated with following formula:
ScoreJj=simJj*SSi
Wherein, ScoreJj is the initial characteristics score value of cell dictionary Jj to be recommended, and simJj is cell dictionary Jj to be recommendedSimilarity, SSi be the corresponding mounted cell dictionary Si of cell dictionary Jj to be recommended feature score value, i=1,2 ... ...,N, N are the quantity of the mounted cell dictionary of active user；J=1,2 ... ..., M, M are the quantity of cell dictionary to be recommended.
For example, corresponding feature score value is SS1 for mounted cell dictionary S1, meanwhile, S1 is corresponding to be recommendedCell dictionary includes J1, J2 and J3, the similarity of cell dictionary J1, J2 and J3 to be recommended be respectively simJ1, simJ2 andSimJ3 can then calculate initial association feature score value of the ScoreJ1=SS1*simJ1 as cell dictionary J1 to be recommended, withInitial association feature score value of the ScoreJ2=SS1*simJ2 as cell dictionary J2 to be recommended, with ScoreJ3=SS1*simJ3Initial association feature score value as cell dictionary J3 to be recommended.
Step 205, the corresponding whole initial association feature score values of each cell dictionary to be recommended are added up, are obtained described eachThe linked character score value of cell dictionary to be recommended；
In embodiments of the present invention, since active user may be equipped with multiple cell dictionaries, it may for example comprise mountedCell dictionary S2 and S3, then the corresponding cell dictionary to be recommended of S1, S2 and S3 may include identical cell dictionary.ExampleSuch as, the corresponding cell dictionary to be recommended of S1 be J1, J2 and J3 when, if the corresponding cell dictionary to be recommended of S2 be J1, J4 and J5,The corresponding cell dictionary to be recommended of S3 is J2, J6 and J7, then there are identical cell dictionary J1 to be recommended, S1 to deposit with S3 with S2 by S1In identical cell dictionary J2 to be recommended.At this point, for duplicate removal each cell dictionary to be recommended can be being calculated separately outAfter initial association feature score value, the initial association feature score value of identical cell dictionary to be recommended is added, as final passJoin feature score value.
For example, the initial association feature point for feature score value J1 to be recommended, relative to mounted cell dictionary S1Value is SS1*simJ1, and the initial association feature score value relative to mounted cell dictionary S2 is SS2*simJ1 (wherein, SS2For the feature score value of S2), then it can linked character score value using SS1*simJ1+SS2*simJ1 as J1.
Step 206, the cell dictionary to be recommended is ranked up according to the linked character score value；
It in the concrete realization, can be special according to association after the linked character score value for obtaining each cell dictionary to be recommendedThe ascending or descending order arrangement of score value is levied, the embodiment of the present invention is not construed as limiting this.
Step 207, the cell dictionary to be recommended for extracting the preceding preset quantity that sorts is target cell dictionary；
In embodiments of the present invention, several highest cell dictionaries of linked character score value can be extracted as target cellDictionary.
Specifically, it if being ranked up according to linked character score value descending to each cell dictionary to be recommended, can chooseSeveral preceding cell dictionaries sort as target cell dictionary.The quantity of target cell dictionary can be by those skilled in the artDetermine according to actual needs, the embodiment of the present invention is not construed as limiting this to member.
Step 208, recommend the target cell dictionary.
In embodiments of the present invention, after determining target cell dictionary, initiatively the target cell dictionary can be recommendedTo user.For example, user installation target cell dictionary can be prompted, or the target cell word is directly installed in input methodLibrary, the embodiment of the present invention are not construed as limiting the concrete mode of recommendation.
In order to make it easy to understand, below with a specific example, to the recommended method of the cell dictionary of the embodiment of the present inventionIt makes a presentation.
(1) determine which the mounted cell dictionary of active user includes first, client can be according to user to eachThe service condition of mounted cell dictionary determines the feature score value of each mounted cell dictionary.Specifically, active userWhen one new cell dictionary of every installation, the feature score value of the cell dictionary is set as 1, if the word shielded on user hits this carefullyBorn of the same parents' dictionary then adds 1 to feature score value, if the word that user deletes hits the cell dictionary, subtracts 1 to feature score value；For working asThe preceding uninstalled cell dictionary of user is then unified to set feature score value as 0.The client every 24 hours spies by whole cell dictionariesSign score value is uploaded to server.
(2) can be in the big matrix of one N*M of server maintenance, the feature point of the cell dictionary for storing user's uploadValue, N is the quantity of all users, and M is the sum of all cell dictionaries.In the matrix, every a line indicates a cell dictionaryCorresponding to the feature score value of all users, each column then indicate that all cell dictionaries correspond to the feature score value of a user.SoAfterwards, using the feature score value of every a line as a vector, N number of feature of the cell dictionary is indicated, using Pearson correlation coefficientThe every 5 days similarities calculated between an any two vector, as the similarity between corresponding two cell dictionaries.
(3) for each mounted cell dictionary, according to the size of similarity, select similarity highest several respectivelyA cell dictionary is as cell dictionary to be recommended, and with the similarity of each cell dictionary to be recommended and corresponding mounted thinInitial association feature score value of the product of the feature score value of born of the same parents' dictionary as each cell dictionary to be recommended.If one to be recommendedCell dictionary corresponds to the mounted cell dictionary of more than one, then whole initial associations of this cell dictionary to be recommended are specialSign score value carries out aggregation as linked character score value, to achieve the purpose that duplicate removal, and using final calculated result as allCell dictionary to be recommended linked character score value.
(4) according to the descending of linked character score value, whole cell dictionaries to be recommended is ranked up, selected and sorted is precedingK cell dictionaries to be recommended be target cell dictionary, and this K target cell dictionary is recommended into user.
It should be noted that for simple description, therefore, it is stated as a series of action groups for embodiment of the methodIt closes, but those skilled in the art should understand that, embodiment of that present invention are not limited by the describe sequence of actions, because according toAccording to the embodiment of the present invention, some steps may be performed in other sequences or simultaneously.Secondly, those skilled in the art also shouldKnow, the embodiments described in the specification are all preferred embodiments, and the related movement not necessarily present invention is implementedNecessary to example.
Referring to Fig. 3, a kind of structural block diagram of the recommendation apparatus embodiment of cell dictionary of the invention is shown, it specifically can be withIncluding following module:
Determining module 301, for determining that mounted cell dictionary, the mounted cell dictionary have corresponding specialLevy score value；
Module 302 is obtained, for obtaining at least one corresponding cell dictionary to be recommended of each mounted cell dictionary,The cell dictionary to be recommended has corresponding similarity with described installed between cell dictionary；
Computing module 303, for the feature score value and the cell word to be recommended using the mounted cell dictionaryThe similarity in library calculates the linked character score value of each cell dictionary to be recommended；
Extraction module 304, for it is thin to extract target from the cell dictionary to be recommended according to the linked character score valueBorn of the same parents' dictionary；
Recommending module 305, for recommending the target cell dictionary.
In embodiments of the present invention, the acquisition module 302 can specifically include following submodule:
Similarity calculation submodule, it is similar between each mounted cell dictionary and other cell dictionaries for calculatingDegree；
Cell dictionary extracting sub-module to be recommended, for extracting at least one cell that the similarity is more than preset thresholdDictionary is as cell dictionary to be recommended.
In embodiments of the present invention, the similarity calculation submodule can specifically include such as lower unit:
User characteristics score value acquiring unit, for obtaining the user characteristics score value of each cell dictionary；
Feature vector generation unit generates each cell for the user characteristics score value according to each cell dictionaryThe feature vector of dictionary；
Similarity calculated calculates each mounted cell dictionary and other is thin for using described eigenvectorSimilarity between born of the same parents' dictionary.
In embodiments of the present invention, the user characteristics score value acquiring unit can specifically include following subelement:
First setting subelement, for setting the initial characteristics score value of each mounted cell dictionary；
It is incremented by subelement, for having been installed to described when the word shielded on user is the word in mounted cell dictionaryThe corresponding initial characteristics score value of cell dictionary be incremented by；
Successively decrease subelement, for having been installed to described when the word that user deletes is the word in mounted cell dictionaryThe corresponding initial characteristics score value of cell dictionary successively decrease；
User characteristics score value determines subelement, is used for revised initial characteristics score value, as described mounted thinThe user characteristics score value of born of the same parents' dictionary.
In embodiments of the present invention, the similarity calculated can specifically include following subelement:
Similarity calculation subelement, for calculating separately the feature vector and other cells of each mounted cell dictionaryThe distance between feature vector of dictionary, using the distance as between each mounted cell dictionary and other cell dictionariesSimilarity.
In embodiments of the present invention, the computing module can specifically include following submodule:
Initial association feature score value computational submodule, for the feature score value of the mounted cell dictionary and describedInitial association feature score value of the similarity product of cell dictionary to be recommended as the cell dictionary to be recommended；
Linked character score value computational submodule, it is special for adding up the corresponding whole initial associations of each cell dictionary to be recommendedScore value is levied, the linked character score value of each cell dictionary to be recommended is obtained.
In embodiments of the present invention, the extraction module can specifically include following submodule:
Sorting sub-module, for being ranked up according to the linked character score value to the cell dictionary to be recommended；
Extracting sub-module, the cell dictionary to be recommended for extracting the preceding preset quantity that sorts are target cell dictionary.
For device embodiment, since it is basically similar to the method embodiment, related so being described relatively simplePlace illustrates referring to the part of embodiment of the method.
Fig. 4 is a kind of block diagram of the recommendation apparatus 400 of cell dictionary shown according to an exemplary embodiment.For example, dressSetting 400 can be mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical treatmentEquipment, body-building equipment, personal digital assistant etc..
Referring to Fig. 4, device 400 may include following one or more components: processing component 402, memory 404, power supplyComponent 406, multimedia component 408, audio component 410, the interface 412 of input/output (I/O), sensor module 414, andCommunication component 416.
The integrated operation of the usual control device 400 of processing component 402, such as with display, telephone call, data communication, phaseMachine operation and record operate associated operation.Processing element 402 may include that one or more processors 420 refer to executeIt enables, to complete all or part of the steps of the recommended method of above-mentioned cell dictionary.In addition, processing component 402 may include oneA or multiple modules, convenient for the interaction between processing component 402 and other assemblies.For example, processing component 402 may include more matchmakersModule, to facilitate the interaction between multimedia component 408 and processing component 402.
Memory 404 is configured as storing various types of data to support the operation in device 400.These data are shownExample includes the instruction of any application or method for operating on device 400, contact data, and telephone book data disappearsBreath, picture, video etc..Memory 404 can be by any kind of volatibility or non-volatile memory device or their groupIt closes and realizes, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM) is erasable to compileJourney read-only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flashDevice, disk or CD.
Power supply module 406 provides electric power for the various assemblies of device 400.Power supply module 406 may include power management systemSystem, one or more power supplys and other with for device 400 generate, manage, and distribute the associated component of electric power.
Multimedia component 408 includes the screen of one output interface of offer between described device 400 and user.OneIn a little embodiments, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screenCurtain may be implemented as touch screen, to receive input signal from the user.Touch panel includes one or more touch sensingsDevice is to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or sliding actionBoundary, but also detect duration and pressure associated with the touch or slide operation.In some embodiments, more matchmakersBody component 408 includes a front camera and/or rear camera.When device 400 is in operation mode, such as screening-mode orWhen video mode, front camera and/or rear camera can receive external multi-medium data.Each front camera andRear camera can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 410 is configured as output and/or input audio signal.For example, audio component 410 includes a MikeWind (MIC), when device 400 is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone is matchedIt is set to reception external audio signal.The received audio signal can be further stored in memory 404 or via communication setPart 416 is sent.In some embodiments, audio component 410 further includes a loudspeaker, is used for output audio signal.
I/O interface 412 provides interface between processing component 402 and peripheral interface module, and above-mentioned peripheral interface module canTo be keyboard, click wheel, button etc..These buttons may include, but are not limited to: home button, volume button, start button and lockDetermine button.
Sensor module 414 includes one or more sensors, and the state for providing various aspects for device 400 is commentedEstimate.For example, sensor module 414 can detecte the state that opens/closes of device 400, and the relative positioning of component, for example, it is describedComponent is the display and keypad of device 400, and sensor module 414 can be with 400 1 components of detection device 400 or devicePosition change, the existence or non-existence that user contacts with device 400,400 orientation of device or acceleration/deceleration and device 400Temperature change.Sensor module 414 may include proximity sensor, be configured to detect without any physical contactPresence of nearby objects.Sensor module 414 can also include optical sensor, such as CMOS or ccd image sensor, atAs being used in application.In some embodiments, which can also include acceleration transducer, gyro sensorsDevice, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 416 is configured to facilitate the communication of wired or wireless way between device 400 and other equipment.Device400 can access the wireless network based on communication standard, such as WiFi, 2G or 3G or their combination.In an exemplary implementationIn example, communication component 416 receives broadcast singal or broadcast related information from external broadcasting management system via broadcast channel.In one exemplary embodiment, the communication component 416 further includes near-field communication (NFC) module, to promote short range communication.ExampleSuch as, NFC module can be based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band (UWB) technology,Bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, device 400 can be believed by one or more application specific integrated circuit (ASIC), numberNumber processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing the recommendation of above-mentioned cell dictionaryMethod.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instruction, example are additionally providedIt such as include the memory 404 of instruction, above-metioned instruction can be executed by the processor 420 of device 400 to complete above-mentioned cell dictionaryRecommended method.For example, the non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and optical data storage devices etc..
A kind of recommendation apparatus of cell dictionary includes memory and one or more than one program, wherein oneA perhaps more than one program is stored in memory and is configured to execute described one by one or more than one processorA or more than one program includes the instruction for performing the following operation: determine mounted cell dictionary, it is described to have installedCell dictionary have corresponding feature score value；Obtain at least one corresponding cell to be recommended of each mounted cell dictionaryDictionary, the cell dictionary to be recommended have corresponding similarity with described installed between cell dictionary；Pacified using describedThe similarity of the feature score value of the cell dictionary of dress and the cell dictionary to be recommended calculates the pass of each cell dictionary to be recommendedJoin feature score value；According to the linked character score value, target cell dictionary is extracted from the cell dictionary to be recommended；Recommend instituteState target cell dictionary.
Optionally, the one or more programs also include the instruction for performing the following operation: being calculated eachSimilarity between mounted cell dictionary and other cell dictionaries；Extract the similarity is more than preset threshold at least oneA cell dictionary is as cell dictionary to be recommended.
Optionally, the one or more programs also include the instruction for performing the following operation: being obtained eachThe user characteristics score value of cell dictionary；According to the user characteristics score value of each cell dictionary, each cell dictionary is generatedFeature vector；Using described eigenvector, the similarity between each mounted cell dictionary and other cell dictionaries is calculated.
Optionally, the one or more programs also include the instruction for performing the following operation: setting is eachThe initial characteristics score value of mounted cell dictionary；When the word shielded on user is the word in mounted cell dictionary, to instituteThe corresponding initial characteristics score value of mounted cell dictionary is stated to be incremented by；When the word that user deletes is in mounted cell dictionaryWhen word, successively decrease to the corresponding initial characteristics score value of the mounted cell dictionary；With revised initial characteristics score value, asThe user characteristics score value of the mounted cell dictionary.
Optionally, the one or more programs also include the instruction for performing the following operation: being calculated separatelyThe feature vector of each mounted cell dictionary and the distance between the feature vector of other cell dictionaries are made with the distanceFor the similarity between each mounted cell dictionary and other cell dictionaries.
Optionally, the one or more programs also include the instruction for performing the following operation: with it is describedThe similarity product of the feature score value of the cell dictionary of installation and the cell dictionary to be recommended is as the cell word to be recommendedThe initial association feature score value in library；The corresponding whole initial association feature score values of each cell dictionary to be recommended are added up, institute is obtainedState the linked character score value of each cell dictionary to be recommended.
Optionally, the one or more programs also include the instruction for performing the following operation: according to described inLinked character score value is ranked up the cell dictionary to be recommended；Extract the cell word to be recommended for the preceding preset quantity that sortsLibrary is target cell dictionary.
A kind of storage medium is enabled the terminal to when the instruction in the storage medium is executed by the processor of terminalIt performs the following operations: determining that mounted cell dictionary, the mounted cell dictionary have corresponding feature score value；It obtainsAt least one corresponding cell dictionary to be recommended of each mounted cell dictionary, the cell dictionary to be recommended have been pacified with describedFilling has corresponding similarity between cell dictionary；Using the feature score value of the mounted cell dictionary and described to be recommendedThe similarity of cell dictionary calculates the linked character score value of each cell dictionary to be recommended；According to the linked character score value, fromTarget cell dictionary is extracted in the cell dictionary to be recommended；Recommend the target cell dictionary.
Optionally, when the instruction in the storage medium is executed by the processor of terminal, terminal can also execute as followsOperation: the similarity between each mounted cell dictionary and other cell dictionaries is calculated；It is more than pre- for extracting the similarityIf at least one cell dictionary of threshold value is as cell dictionary to be recommended.
Optionally, when the instruction in the storage medium is executed by the processor of terminal, terminal can also execute as followsOperation: the user characteristics score value of each cell dictionary is obtained；According to the user characteristics score value of each cell dictionary, generate everyThe feature vector of a cell dictionary；Using described eigenvector, each mounted cell dictionary and other cell dictionaries are calculatedBetween similarity.
Optionally, when the instruction in the storage medium is executed by the processor of terminal, terminal can also execute as followsOperation: the initial characteristics score value of each mounted cell dictionary of setting；When the word shielded on user is mounted cell dictionaryIn word when, the corresponding initial characteristics score value of the mounted cell dictionary is incremented by；When the word that user deletes is to have installedCell dictionary in word when, successively decrease to the corresponding initial characteristics score value of the mounted cell dictionary；With revised firstBeginning feature score value, the user characteristics score value as the mounted cell dictionary.
Optionally, when the instruction in the storage medium is executed by the processor of terminal, terminal can also execute as followsOperation: calculate separately between the feature vector of each mounted cell dictionary and the feature vector of other cell dictionaries away fromFrom using the distance as the similarity between each mounted cell dictionary and other cell dictionaries.
Optionally, when the instruction in the storage medium is executed by the processor of terminal, terminal can also execute as followsOperation: using the similarity product of the feature score value of the mounted cell dictionary and the cell dictionary to be recommended described inThe initial association feature score value of cell dictionary to be recommended；Add up the corresponding whole initial association features of each cell dictionary to be recommendedScore value obtains the linked character score value of each cell dictionary to be recommended.
Optionally, when the instruction in the storage medium is executed by the processor of terminal, terminal can also execute as followsOperation: the cell dictionary to be recommended is ranked up according to the linked character score value；Extract the preceding preset quantity that sortsCell dictionary to be recommended be target cell dictionary.
All the embodiments in this specification are described in a progressive manner, the highlights of each of the examples are withThe difference of other embodiments, the same or similar parts between the embodiments can be referred to each other.
It should be understood by those skilled in the art that, the embodiment of the embodiment of the present invention can provide as method, apparatus or calculateMachine program product.Therefore, the embodiment of the present invention can be used complete hardware embodiment, complete software embodiment or combine software andThe form of the embodiment of hardware aspect.Moreover, the embodiment of the present invention can be used one or more wherein include computer canWith in the computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) of program codeThe form of the computer program product of implementation.
The embodiment of the present invention be referring to according to the method for the embodiment of the present invention, terminal device (system) and computer programThe flowchart and/or the block diagram of product describes.It should be understood that flowchart and/or the block diagram can be realized by computer program instructionsIn each flow and/or block and flowchart and/or the block diagram in process and/or box combination.It can provide theseComputer program instructions are set to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing terminalsStandby processor is to generate a machine, so that being held by the processor of computer or other programmable data processing terminal devicesCapable instruction generates for realizing in one or more flows of the flowchart and/or one or more blocks of the block diagramThe device of specified function.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing terminal devicesIn computer-readable memory operate in a specific manner, so that instruction stored in the computer readable memory generates packetThe manufacture of command device is included, which realizes in one side of one or more flows of the flowchart and/or block diagramThe function of being specified in frame or multiple boxes.
These computer program instructions can also be loaded into computer or other programmable data processing terminal devices, so thatSeries of operation steps are executed on computer or other programmable terminal equipments to generate computer implemented processing, thusThe instruction executed on computer or other programmable terminal equipments is provided for realizing in one or more flows of the flowchartAnd/or in one or more blocks of the block diagram specify function the step of.
Although the preferred embodiment of the embodiment of the present invention has been described, once a person skilled in the art knows basesThis creative concept, then additional changes and modifications can be made to these embodiments.So the following claims are intended to be interpreted asIncluding preferred embodiment and fall into all change and modification of range of embodiment of the invention.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to byOne entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operationBetween there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaningCovering non-exclusive inclusion, so that process, method, article or terminal device including a series of elements not only wrapThose elements are included, but also including other elements that are not explicitly listed, or further includes for this process, method, articleOr the element that terminal device is intrinsic.In the absence of more restrictions, being wanted by what sentence "including a ..." limitedElement, it is not excluded that there is also other identical elements in process, method, article or the terminal device for including the element.
Recommended method to a kind of cell dictionary provided by the present invention and a kind of recommendation apparatus of cell dictionary above, intoIt has gone and has been discussed in detail, used herein a specific example illustrates the principle and implementation of the invention, the above implementationThe explanation of example is merely used to help understand method and its core concept of the invention；Meanwhile for the general technology people of this fieldMember, according to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion this explanationBook content should not be construed as limiting the invention.