CN108710654B - Public opinion data visualization method and equipment - Google Patents

Abstract

The utility model provides a public sentiment data visualization method and equipment, the scheme is after gathering the public sentiment data, confirm the emotion classification and the time information of every public sentiment data earlier, then according to the emotion classification and the time information of public sentiment data, confirm the quantity of the public sentiment data that belong to the target emotion classification in every preset time quantum, and obtain the keyword of the public sentiment data that belong to the target emotion classification in every preset time quantum, and then according to the quantity of the public sentiment data that belong to the target emotion classification in every preset time quantum, generate the visual figure about every preset time quantum, and add corresponding keyword in the visual figure. Since the emotional tendency, the quantity and the related keywords of the public opinion data in the preset time period are intuitively represented in the visual graph, the user can comprehensively and efficiently know the public opinion condition in the preset time period by looking up the visual graph.

Description

Public opinion data visualization method and equipment

Technical Field

The application relates to the technical field of information, in particular to a public opinion data visualization method and device.

Background

Under the big data era of the internet, various new media forms such as media are developed, aiming at various comments and articles of social events and news reports are good and irregular, fishes and dragons are mixed, and more useful and malicious injuries and rumors are caused, false information is released deliberately, or for unreal reports, wave-pushing and help billows are attempted, and the misleading public opinion is spread by means of the fast and various internet, so that large-scale public sentiment events can occur, and huge losses are caused to the country and the society. At present, public opinion data such as news reports, event messages and comments on the internet have a plurality of publishing channels and huge information amount, so that a mode which can enable a user to comprehensively and efficiently know the current public opinion tendency does not exist.

Disclosure of Invention

An object of the present application is to provide a public opinion data visualization scheme, which can let users comprehensively and efficiently know the current public opinion tendency, and then achieve the monitoring of public opinion.

In order to achieve the above object, the present application provides a public opinion data visualization method, which includes:

public opinion data is collected;

determining the emotion category and time information of each piece of public opinion data;

determining the quantity of the public sentiment data belonging to the target sentiment category in each preset time period according to the sentiment category and the time information of the public sentiment data, and acquiring keywords of the public sentiment data belonging to the target sentiment category in each preset time period;

and generating a visual graph related to each preset time period according to the quantity of the public sentiment data belonging to the target sentiment category in each preset time period, and adding corresponding keywords in the visual graph.

Based on another aspect of this application, still provide a public opinion data visualization device, wherein, this equipment includes:

the data acquisition device is used for acquiring public opinion data;

the data processing device is used for determining the emotion type and time information of each piece of public sentiment data, determining the quantity of the public sentiment data belonging to the target emotion type in each preset time period according to the emotion type and time information of the public sentiment data, and acquiring the keywords of the public sentiment data belonging to the target emotion type in each preset time period;

and the data coding device is used for generating a visual graph related to each preset time period according to the quantity of the public sentiment data belonging to the target sentiment category in each preset time period, and adding corresponding keywords in the visual graph.

In addition, this application still provides a public opinion data visualization equipment, and this equipment includes:

a processor; and

one or more machine-readable media storing machine-readable instructions that, when executed by the processor, cause the apparatus to perform the aforementioned public opinion data visualization method.

In the scheme that this application provided, after gathering public opinion data, confirm the emotion classification and the time information of every public opinion data earlier, then according to the emotion classification and the time information of public opinion data, the quantity of the public opinion data who belongs to target emotion classification in the definite preset time quantum, and acquire the keyword of the public opinion data who belongs to target emotion classification in the preset time quantum, and then according to the quantity of the public opinion data who belongs to target emotion classification in the preset time quantum, generate about the visual figure of preset time quantum, and add corresponding keyword in the visual figure. Since the emotional tendency, the quantity and the related keywords of the public opinion data in the preset time period are intuitively represented in the visual graph, the user can comprehensively and efficiently know the public opinion condition in the preset time period by looking up the visual graph.

Drawings

Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:

fig. 1 is a processing flow chart of a public opinion data visualization method according to an embodiment of the present application;

FIG. 2 is a schematic illustration of a visualization layout involved in an embodiment of the present application;

fig. 3 is an overall flowchart of the public opinion data visualization graph generated by using the scheme provided in the embodiment of the present application;

FIG. 4 shows a detailed processing flow of data processing steps in an embodiment of the present application;

FIG. 5 shows a detailed process flow of the data encoding step in one embodiment of the present application;

fig. 6 is a schematic structural diagram of a public opinion data visualization device according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of another public opinion data visualization device according to an embodiment of the present application;

the same or similar reference numbers in the drawings identify the same or similar elements.

Detailed Description

The present application is described in further detail below with reference to the attached figures.

In a typical configuration of the present application, the terminal, the devices serving the network each include one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, which include both non-transitory and non-transitory, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, program means, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

The embodiment of the application provides a public opinion data visualization method, which can generate a visual graph according to public opinion data collected in a period of time, and visually express emotional tendency, quantity and related keywords of the public opinion data in a preset time period in the visual graph, so that a user can comprehensively and efficiently know the public opinion in a certain time period by looking up the visual graph. In a practical scenario, the execution subject of the method may include, but is not limited to, a network host, a single network server, a plurality of network server sets, or a cloud computing-based computer collection, etc. Here, the Cloud is made up of a large number of hosts or web servers based on Cloud Computing (Cloud Computing), which is a type of distributed Computing, one virtual computer consisting of a collection of loosely coupled computers.

Fig. 1 shows a processing flow of a public opinion data visualization method provided by an embodiment of the application, and the processing flow includes the following processing steps:

and S101, collecting public opinion data. In the implementation of the application, the public opinion data is data related to public opinion conditions, can be derived from various media such as news reports, forums, blogs, microblogs, community comments and the like, and can reflect attitudes and emotional tendencies of social public to occurrence, development and changes of social events. In the internet scenario, a crawler (web crawler) may be used to obtain various public opinion data from the internet according to preset rules, for example, periodically obtain various news and comments on the news from various web portals.

S102, determining the emotion type and time information of each piece of public opinion data for classifying the public opinion data, and taking the public opinion data of the same emotion type in a specific time period as a set.

When determining the emotion type of public opinion data, the emotion type of each public opinion data can be identified and marked according to the content of each public opinion data. The emotion categories can be set according to actual classification requirements, for example, in the embodiment of the application, the emotion categories are set to be two categories, namely, positive emotion and negative emotion, and the positive emotion and the negative emotion can be further subdivided in an actual scene according to the actual requirements, so that more emotion categories can be obtained. In the embodiment of the application, if there are 1 to N pieces of public sentiment data collected this time, after classification is performed according to the positive and negative of sentiment, 1 to M pieces of public sentiment data are classified as positive sentiment, and marks from M +1 to N are classified as negative sentiment.

The emotion classification of the public sentiment data can be determined based on the content of the public sentiment data, for example, if the content of a certain news is an event reporting the sex of a certain place and affirming the behavior, the emotion classification of the news is classified as positive emotion. In an actual scene, a machine learning algorithm can be used for automatic recognition, firstly, a training set marked with emotion classes is used for training a machine-learned classification model, and after training is completed, the classification model can recognize the emotion classes of newly-input public opinion data. The specific algorithm of machine learning can be selected according to the requirements of the actual scene, such as algorithms of logistic regression, decision trees, naive Bayes and the like.

When determining the time information of each piece of public opinion data, the publishing time can be determined as the time information of the public opinion data, for example, if the publishing time of a certain public opinion data is 2018-4-22,20:22:22, the time information can be determined as the publishing time. However, in some special cases, for example, if the website does not open a corresponding interface or record the publishing time, the publishing time cannot be obtained, and at this time, the time information of the public opinion data may be determined according to the time of collecting the public opinion data. For example, for a website with a fast update speed, the crawler program is configured to acquire the public opinion data of the website once every 10s, and the time of acquiring the public opinion data each time is set as the time information of the public opinion data. Therefore, in some embodiments of the present application, if public opinion data includes a publishing time, the publishing time is determined as time information of the public opinion data; and if the public opinion data does not contain the publishing time, determining the time information of the public opinion data according to the time for collecting the public opinion data.

Step S103, determining the quantity of the public sentiment data belonging to the target sentiment category in each preset time period according to the sentiment category and the time information of the public sentiment data, and acquiring the keywords of the public sentiment data belonging to the target sentiment category in each preset time period.

If the preset time period is used as a time unit for generating the visual graph, for example, the preset time period is 1 hour, the scheme of the embodiment of the application counts the number of public sentiment data of a specific sentiment category within 1 hour by taking 1 hour as a statistical time interval during processing. For example, the collected public opinion data may include data in the last 5 hours, and at this time, the collected public opinion data is divided into 5 preset time periods according to the time interval of 1 hour, and the number of the public opinion data and the keywords in each preset time period are counted. The target emotion category refers to an emotion category which needs to be embodied in a visual graph, and may include all emotion categories, or may select some emotion categories which need to be paid attention by the user from all emotion categories, for example, in this embodiment of the present application, the target emotion category may be a positive emotion and a negative emotion. Taking a preset time period 00:00:01-01:00:00 as an example, if there are 200 pieces of collected time information of the public sentiment data in the preset time period, wherein there are 130 pieces of positive sentiments in the sentiment categories and 70 pieces of negative sentiments in the sentiment categories, the number of the public sentiment data belonging to the target sentiment categories in the preset time period 00:00:01-01:00:00 can be determined as follows: 130 positive emotions and 70 negative emotions.

For the collected public opinion data, keywords can be extracted by adopting a keyword extraction algorithm. The algorithm used in the keyword extraction in the embodiment of the application can adopt any algorithm suitable for public sentiment data processing scenes, such as TF-IDF, KEA and other algorithms. For the extracted keywords, mark information can be added to the keywords to distinguish the keywords with the same content but from public opinion data in different preset time periods. For example, forpublic opinion data 1, akeyword 1 and a keyword 2 may be extracted, and at the time of recording, flag information, such as attaching a timestamp or other information that can be used to distinguish a preset time period to which the keyword belongs, may be added to the keyword. In one embodiment of the present application, the following form may be used for recording: keyword _ slot _ emotion category, such as keyword 1_ slot 1_ positive, keyword 1_ slot 2_ positive, and so on. The time period is a preset time period to which the public sentiment data corresponding to thekeyword 1 belongs, and the emotion category is an emotion category of the public sentiment data corresponding to thekeyword 1.

And step S104, generating a visual graph related to each preset time period according to the quantity of the public sentiment data belonging to the target sentiment category in each preset time period, and adding corresponding keywords in the visual graph.

When generating the visual pattern for each preset time period, a pattern attribute of the visual pattern for each preset time period may be determined according to the number of public sentiment data belonging to the target sentiment category in each preset time period, where the pattern attribute may be a parameter related to a visual image of the visual pattern, and may be specifically determined according to an actually adopted pattern, for example, any one or a combination of multiple items of an area, a height, a width, a diameter, and a curvature of the visual pattern. After determining the graphical attributes, the visualization may be generated within the presentation area of the target emotion classification based on the graphical attributes. For example, when the visualization graph is a bar graph, the graphic attribute may be the height of the bar graph, and when the visualization graph is a line graph, the graphic attribute may be the area of a closed graph surrounded by the line graph, or the like.

In some embodiments of the present application, before generating a visual figure for each preset time period according to the number of public sentiment data belonging to the target sentiment category in each preset time period, a visual layout may be determined, wherein the visual layout at least includes a display area of the target sentiment category and a form of the visual figure. For example, fig. 2 shows a visualization layout in the embodiment of the present application, which adopts a mirror image layout style, that is, a horizontal axis is taken as time, which is used as a baseline for dividing display areas of two emotion categories, the display areas of two emotion categories are divided up and down, a display area of positive emotion is above the baseline, a display area of negative emotion is below the baseline, the number of public opinion data belonging to a target emotion category in a preset time period is taken as a vertical axis, and a form of a visualization graph is a line graph.

In some embodiments of the application, when corresponding keywords are added to the visual graph, the word frequency of the keywords of the public sentiment data belonging to the target emotion category in each preset time period may be counted first, and then the corresponding keywords are added to the visual graph in a word cloud form according to the word frequency of the keywords. By adopting the form of word cloud, the display size of the keywords in the visual graph is related to the word frequency of the keywords, for example, the display size of the keywords with high word frequency is larger, and the display size of the keywords with low word frequency is relatively smaller.

In the embodiment of the application, after the keyword is extracted, the statistical information of the keyword can be recorded in a form of 'keyword _ time period _ emotion category'. According to the statistical information, the word frequency of the key words in each time period can be further counted, for example, for thekey word 1, the statistical information has 60 records of thekey word 1 in total, wherein 10 records of the key word 1_ time period 1_ positive ", 20 records of the key word 1_ time period 2_ positive", and 30 records of the key word 1_ time period 4_ positive ". In constructing the word cloud, the size of the display size is related to the word frequency of occurrence of the keyword, so that the display size of the keyword in each preset time period can be determined proportionally, for example, for theaforementioned keyword 1, it appears 10 times in thepreset time period 1, 20 times in the preset time period 2, and 30 times in thepreset time period 4, so that when the word cloud is generated, the display size is the smallest in thepreset time period 1, the second in the preset time period 2, and the display size is the largest in thepreset time period 4.

Further, word frequencies of other keywords can be determined in the same manner, and then the display sizes of different keywords in the word cloud in different preset time periods are determined uniformly based on the word frequencies of all the keywords.

In addition, since the visual pattern corresponds to a preset time period and the display position of the visual pattern is related to the time, when the keyword is added, the display position of the keyword can also be related to the time, and the user can obtain more information from the visual pattern to which the keyword is added, for example, the user can determine the topic mainly concerned by the public sentiment at a certain time through the keyword at the position corresponding to the certain time. Therefore, when a corresponding keyword is added to the visual graph, the position information of the keyword in the visual graph can be determined according to the time information of the keyword, and then the corresponding keyword is added to the corresponding position in the visual graph according to the position information, wherein the time information of the keyword is the time information of public sentiment data to which the keyword belongs.

Further, in some embodiments of the application, colors can be added to the visual graphics and the keywords, and the colors are related to the emotion categories corresponding to the visual graphics, so that the user can feel the overall situation of the public opinion data of each emotion category more intuitively from the aspect of colors. In an actual scene, the selection of colors may be set according to the usage habits of the user, for example, for the two emotion categories in the embodiment of the present application, the colors of warm color systems such as red and yellow may be used to mark positive emotions, and the colors of cold color systems such as blue and cyan may be used to mark negative emotions.

Fig. 3, fig. 4 and fig. 5 show a processing flow when generating a public opinion data visualization graph by using the scheme provided by the embodiment of the application, and the processing flow comprises the following processing steps:

and step S1, data acquisition.

Step S2, data processing.

Step S2 includes 3 parts of content, as shown in fig. 4. Marking and classifying collected public sentiment data according to positive and negative sentiments, marking and classifying thecontent data 1 to M into positive sentiment data according to the positive and negative sentiments on the assumption that the data contains 1 to N pieces of data, and marking and classifying thecontent data 1 to M into negative sentiment data according to the positive and negative sentiments. Secondly, setting a time interval as a preset time period, such as 1 hour, and then carrying out public opinion data quantity statistics in one time period by one time period. And finally, in each preset time period, counting the word frequency of the keywords appearing in the preset time period according to the positive and negative classification of the emotion types.

Step S3, data encoding.

Step S3 is shown in fig. 5 as the following 3 parts. First, a visualization layout of a visualization graph for data presentation is determined. The mirror image layout is used as a layout style, namely time is used as a horizontal axis, the number of public sentiment data is used as a vertical axis, the number of the public sentiment data is 0 as a base line, a display area of positive sentiment is arranged above the base line, and a display area of negative sentiment is arranged below the base line.

Secondly, coding the quantity of the public sentiment data according to the sentiment category and the determined time period, and coding the public sentiment data in a line graph form according to the counted quantity in the sentiment positive and negative regions time period by time period respectively. The coding form can be a line graph or other forms which can form a closed area with the base line, such as a graph, and the size of the area formed by the line graph or the graph and the base line represents the number, the positive and negative of the emotion types can be coded by different colors for distinguishing, for example, red can be used for coding negative emotions, and blue can be used for coding positive emotions.

And finally, coding the keywords, carrying out word cloud coding on the processed keywords and word frequency data in each time period according to the emotion types, displaying the keyword data in a word cloud form, wherein the word frequency data of the keywords are coded according to the sizes of words, the spatial positions of the keywords are determined according to the emotion types and the time information of the emotion types, and the colors of the keywords can be coded according to the positive and negative of the public sentiments by using the color tones with the same color as the public sentiments of the keywords to indicate the emotion types.

And step S4, visually outputting. The data processed and coded according to the process is displayed in a visual graph form by using a computer or other means, so that a set of visual display scheme aiming at the multi-dimensional time sequence data coupled with the public opinion data emotion change trend data, the quantity, the public opinion keywords and the word frequency data can be obtained, the scheme can help to master the overall development situation of the hot public opinions and events and the main comments, opinions, viewpoints and emotional tendencies of the public and netizens on the hot public opinions and events, further the monitoring and management on the internet public opinions can be achieved, and the visual display scheme can also be used as the basis for analyzing and judging the public opinion events.

Based on the same inventive concept, the embodiment of the present application further provides a public sentiment data visualization device, the corresponding method of the device is the public sentiment data visualization method in the foregoing embodiment, and the principle of solving the problem is similar to that of the method.

The public opinion data visualization equipment provided by the embodiment of the application can generate the visual graph according to the public opinion data collected in a period of time, and intuitively express the emotional tendency, the quantity and the related keywords of the public opinion data in the preset time period in the visual graph, so that a user can comprehensively and efficiently know the public opinion condition in a certain time period by looking over the visual graph. In a practical scenario, the specific implementation of the apparatus may include, but is not limited to, a network host, a single network server, multiple network server sets, or a cloud computing-based computer set. Here, the Cloud is made up of a large number of hosts or web servers based on Cloud Computing (Cloud Computing), which is a type of distributed Computing, one virtual computer consisting of a collection of loosely coupled computers.

Fig. 6 shows a structure of public opinion data visualization equipment provided by an embodiment of the present application, which includes adata collecting device 610, adata processing device 620, and adata encoding device 630. Wherein, thedata acquisition device 610 is used for gathering public opinion data. In the implementation of the application, the public opinion data is data related to public opinion conditions, can be derived from various media such as news reports, forums, blogs, microblogs, community comments and the like, and can reflect attitudes and emotional tendencies of social public to occurrence, development and changes of social events. In the internet scenario, a crawler (web crawler) may be used to obtain various public opinion data from the internet according to preset rules, for example, periodically obtain various news and comments on the news from various web portals.

Thedata processing device 620 is used for determining the emotion type and time information of each piece of public opinion data, so as to classify the public opinion data, and the public opinion data of the same emotion type in a specific time period is used as a set.

When determining the emotion classification of public opinion data, the data processing device can identify and mark the emotion classification of each public opinion data according to the content of each public opinion data. The emotion categories can be set according to actual classification requirements, for example, in the embodiment of the application, the emotion categories are set to be two categories, namely, positive emotion and negative emotion, and the positive emotion and the negative emotion can be further subdivided in an actual scene according to the actual requirements, so that more emotion categories can be obtained. In the embodiment of the application, if there are 1 to N pieces of public sentiment data collected this time, after classification is performed according to the positive and negative of sentiment, 1 to M pieces of public sentiment data are classified as positive sentiment, and marks from M +1 to N are classified as negative sentiment.

The emotion classification of the public sentiment data can be determined based on the content of the public sentiment data, for example, if the content of a certain news is an event reporting the sex of a certain place and affirming the behavior, the emotion classification of the news is classified as positive emotion. In an actual scene, a machine learning algorithm can be used for automatic recognition, firstly, a training set marked with emotion classes is used for training a machine-learned classification model, and after training is completed, the classification model can recognize the emotion classes of newly-input public opinion data. The specific algorithm of machine learning can be selected according to the requirements of the actual scene, such as algorithms of logistic regression, decision trees, naive Bayes and the like.

In determining the time information of each piece of public opinion data, the data processing apparatus may determine the distribution time as the time information of the public opinion data, for example, if the distribution time of a certain public opinion data is 2018-4-22,20:22:22, the time information thereof may be determined as the distribution time. However, in some special cases, for example, if the website does not open a corresponding interface or record the publishing time, the publishing time cannot be obtained, and at this time, the time information of the public opinion data may be determined according to the time of collecting the public opinion data. For example, for a website with a fast update speed, the crawler program is configured to acquire the public opinion data of the website once every 10s, and the time of acquiring the public opinion data each time is set as the time information of the public opinion data. Therefore, in some embodiments of the present application, if public opinion data includes a publishing time, the publishing time is determined as time information of the public opinion data; and if the public opinion data does not contain the publishing time, determining the time information of the public opinion data according to the time for collecting the public opinion data.

Thedata processing device 620 is further configured to determine, according to the emotion types and the time information of the public opinion data, the number of the public opinion data belonging to the target emotion type in each preset time period, and obtain the keyword of the public opinion data belonging to the target emotion type in each preset time period.

If the preset time period is used as a time unit for generating the visual graph, for example, the preset time period is 1 hour, the scheme of the embodiment of the application counts the number of public sentiment data of a specific sentiment category within 1 hour by taking 1 hour as a statistical time interval during processing. For example, the collected public opinion data may include data in the last 5 hours, and at this time, the collected public opinion data is divided into 5 preset time periods according to the time interval of 1 hour, and the number of the public opinion data and the keywords in each preset time period are counted. The target emotion category refers to an emotion category which needs to be embodied in a visual graph, and may include all emotion categories, or may select some emotion categories which need to be paid attention by the user from all emotion categories, for example, in this embodiment of the present application, the target emotion category may be a positive emotion and a negative emotion. Taking a preset time period 00:00:01-01:00:00 as an example, if there are 200 pieces of collected time information of the public sentiment data in the preset time period, wherein there are 130 pieces of positive sentiments in the sentiment categories and 70 pieces of negative sentiments in the sentiment categories, the number of the public sentiment data belonging to the target sentiment categories in the preset time period 00:00:01-01:00:00 can be determined as follows: 130 positive emotions and 70 negative emotions.

For the collected public opinion data, the data processing device can adopt a keyword extraction algorithm to extract keywords. The algorithm used in the keyword extraction in the embodiment of the application can adopt any algorithm suitable for public sentiment data processing scenes, such as TF-IDF, KEA and other algorithms. For the extracted keywords, mark information can be added to the keywords to distinguish the keywords with the same content but from different public opinion data. For example, forpublic opinion data 1, akeyword 1 and a keyword 2 may be extracted, and at the time of recording, flag information, such as attaching a timestamp or other information that can be used to distinguish a preset time period to which the keyword belongs, may be added to the keyword. In one embodiment of the present application, the following form may be used for recording: keyword _ slot _ emotion category, such as keyword 1_ slot 1_ positive, keyword 1_ slot 2_ positive, and so on. The time period is a preset time period to which the public sentiment data corresponding to thekeyword 1 belongs, and the emotion category is an emotion category of the public sentiment data corresponding to thekeyword 1.

Thedata encoding device 630 is configured to generate a visual graph related to each preset time period according to the number of public sentiment data belonging to the target sentiment category in each preset time period, and add a corresponding keyword in the visual graph.

In generating the visual pattern for each preset time period, the data encoding device may determine a pattern attribute of the visual pattern for each preset time period according to the number of public sentiment data belonging to the target sentiment category in each preset time period, where the pattern attribute may be a parameter related to a visual image of the visual pattern, and may specifically be determined according to a pattern actually adopted, for example, any one or a combination of multiple items of an area, a height, a width, a diameter, and a curvature of the visual pattern. After determining the graphical attributes, the visualization may be generated within the presentation area of the target emotion classification based on the graphical attributes. For example, when the visualization graph is a bar graph, the graphic attribute may be the height of the bar graph, and when the visualization graph is a line graph, the graphic attribute may be the area of a closed graph surrounded by the line graph, or the like.

In some embodiments of the present application, before generating a visual pattern related to a preset time period according to the amount of public sentiment data belonging to a target sentiment category within the preset time period, the data encoding device may determine a visual layout, wherein the visual layout at least includes a display area of the target sentiment category and a form of the visual pattern. For example, fig. 2 shows a visualization layout in the embodiment of the present application, which adopts a mirror image layout style, that is, a horizontal axis is taken as time, which is used as a baseline for dividing display areas of two emotion categories, the display areas of two emotion categories are divided up and down, a display area of positive emotion is above the baseline, a display area of negative emotion is below the baseline, the number of public opinion data belonging to a target emotion category in a preset time period is taken as a vertical axis, and a form of a visualization graph is a line graph.

In some embodiments of the application, when adding corresponding keywords to the visual pattern, the data processing device may count word frequencies of the keywords of the public sentiment data belonging to the target sentiment category in each preset time period, and then the data encoding device adds corresponding keywords to the visual pattern in a word cloud form according to the word frequencies of the keywords. By adopting the form of word cloud, the display size of the keywords in the visual graph is related to the word frequency of the keywords, for example, the display size of the keywords with high word frequency is larger, and the display size of the keywords with low word frequency is relatively smaller.

In the embodiment of the application, after the keyword is extracted, the statistical information of the keyword can be recorded in a form of 'keyword _ time period _ emotion category'. According to the statistical information, the word frequency of the key words in each time period can be further counted, for example, for thekey word 1, the statistical information has 60 records of thekey word 1 in total, wherein 10 records of the key word 1_ time period 1_ positive ", 20 records of the key word 1_ time period 2_ positive", and 30 records of the key word 1_ time period 4_ positive ". In constructing the word cloud, the size of the display size is related to the word frequency of occurrence of the keyword, so that the display size of the keyword in each preset time period can be determined proportionally, for example, for theaforementioned keyword 1, it appears 10 times in thepreset time period 1, 20 times in the preset time period 2, and 30 times in thepreset time period 4, so that when the word cloud is generated, the display size is the smallest in thepreset time period 1, the second in the preset time period 2, and the display size is the largest in thepreset time period 4.

Further, word frequencies of other keywords can be determined in the same manner, and then the display sizes of different keywords in the word cloud in different preset time periods are determined uniformly based on the word frequencies of all the keywords.

In addition, since the visual pattern corresponds to a preset time period and the display position of the visual pattern is related to the time, when the keyword is added, the display position of the keyword can also be related to the time, and the user can obtain more information from the visual pattern to which the keyword is added, for example, the user can determine the topic mainly concerned by the public sentiment at a certain time through the keyword at the position corresponding to the certain time. Therefore, when a corresponding keyword is added to the visual graph, the data coding device may further determine the position information of the keyword in the visual graph according to the time information of the keyword, and then add the corresponding keyword at a corresponding position in the visual graph according to the position information, wherein the time information of the keyword is the time information of the public opinion data to which the keyword belongs.

Further, in some embodiments of the present application, the data encoding device may further add a color to the visual graphics and the keyword, and the color is related to the emotion category corresponding to the visual graphics, so that the user may feel the overall situation of the public opinion data of each emotion category more intuitively from the aspect of color. In an actual scene, the selection of colors may be set according to the usage habits of the user, for example, for the two emotion categories in the embodiment of the present application, the colors of warm color systems such as red and yellow may be used to mark positive emotions, and the colors of cold color systems such as blue and cyan may be used to mark negative emotions.

In addition, some of the present application may be implemented as a computer program product, such as computer program instructions, which when executed by a computer, may invoke or provide methods and/or techniques in accordance with the present application through the operation of the computer. Program instructions which invoke the methods of the present application may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the present application includes an apparatus as shown in fig. 7, which includes one or more machine-readable media 710 storing machine-readable instructions and aprocessor 720 for executing the machine-readable instructions, wherein the machine-readable instructions, when executed by the processor, cause the apparatus to perform the methods and/or aspects according to the embodiments of the present application.

Furthermore, some embodiments of the present application also provide a computer readable medium, on which computer program instructions are stored, the computer readable instructions being executable by a processor to implement the methods and/or aspects of the foregoing embodiments of the present application.

It should be noted that the present application may be implemented in software and/or a combination of software and hardware, for example, implemented using Application Specific Integrated Circuits (ASICs), general purpose computers or any other similar hardware devices. In one embodiment, the software programs of the present application may be executed by a processor to implement the above steps or functions. Likewise, the software programs (including associated data structures) of the present application may be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Additionally, some of the steps or functions of the present application may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.

It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Claims (15)

1. A public opinion data visualization method is provided, wherein the method comprises the following steps:

public opinion data is collected;

determining the emotion category and time information of each piece of public opinion data;

determining the quantity of the public sentiment data belonging to the target sentiment category in each preset time period according to the sentiment category and the time information of the public sentiment data, and acquiring keywords of the public sentiment data belonging to the target sentiment category in each preset time period;

generating a visual graph related to each preset time period according to the quantity of public sentiment data belonging to the target sentiment category in each preset time period, and adding corresponding keywords in the visual graph;

adding corresponding keywords in the visual graph, wherein the adding comprises the following steps:

counting word frequency of keywords of the public sentiment data belonging to the target sentiment category in each preset time period;

and adding corresponding keywords in the visual graph in a word cloud form according to the word frequency of the keywords, wherein the display size of the keywords is related to the word frequency of the keywords.

2. The method of claim 1, wherein determining an emotion classification for each piece of public opinion data comprises:

and identifying and marking the emotion category of each piece of public opinion data according to the content of each piece of public opinion data, wherein the emotion category comprises positive emotion and negative emotion.

3. The method of claim 1, wherein determining time information of each piece of public opinion data comprises:

if public opinion data comprises release time, determining the release time as time information of the public opinion data;

and if the public opinion data does not contain the publishing time, determining the time information of the public opinion data according to the time for collecting the public opinion data.

4. The method of claim 1, wherein generating a visual figure about each preset time period according to the number of public sentiment data belonging to the target sentiment category in each preset time period comprises:

determining the graphic attribute of the visual graphics of each preset time period according to the quantity of the public sentiment data belonging to the target sentiment category in each preset time period;

and generating the visual graph in the display area of the target emotion category according to the graph attribute.

5. The method of claim 4, wherein before generating the visual figure about each preset time period according to the amount of the public sentiment data belonging to the target sentiment category in each preset time period, further comprising:

determining a visualization layout, wherein the visualization layout at least comprises a presentation area of the target emotion category and a form of a visualization graph.

6. The method of claim 1, wherein adding corresponding keywords in the visualization graph comprises:

determining the position information of the keyword in the visual graph according to the time information of the keyword, wherein the time information of the keyword is the time information of public opinion data to which the keyword belongs;

and adding corresponding keywords at corresponding positions in the visual graph according to the position information.

7. The method of claim 1, wherein the method further comprises:

adding colors to the visual graphs and the keywords, wherein the colors are related to the emotion types corresponding to the visual graphs.

8. A public opinion data visualization device, wherein the device comprises:

the data acquisition device is used for acquiring public opinion data;

the data processing device is used for determining the emotion type and time information of each piece of public sentiment data, determining the quantity of the public sentiment data belonging to the target emotion type in each preset time period according to the emotion type and time information of the public sentiment data, acquiring the keywords of the public sentiment data belonging to the target emotion type in each preset time period, and counting the word frequency of the keywords of the public sentiment data belonging to the target emotion type in each preset time period;

and the data coding device is used for generating a visual graph related to each preset time period according to the quantity of the public sentiment data belonging to the target sentiment category in each preset time period, and adding corresponding keywords in the visual graph in a word cloud form according to the word frequency of the keywords, wherein the display size of the keywords is related to the word frequency of the keywords.

9. The device of claim 8, wherein the data processing device is used for identifying and marking the emotion classification of each piece of public opinion data according to the content of each piece of public opinion data, wherein the emotion classification comprises positive emotion and negative emotion.

10. The apparatus according to claim 8, wherein the data processing means determines, when a publication time is included in public opinion data, the publication time as time information of the public opinion data; and when public opinion data does not contain publishing time, determining the time information of the public opinion data according to the time for collecting the public opinion data.

11. The apparatus of claim 8, wherein the data encoding means is configured to determine a graphic attribute of the visual figure for each preset time period according to the number of public sentiment data belonging to the target sentiment category in each preset time period, and generate the visual figure in the display area of the target sentiment category according to the graphic attribute.

12. The apparatus of claim 11, wherein the data encoding device is further configured to determine a visualization layout before generating a visualization graph for each preset time period according to the number of public sentiment data belonging to the target sentiment category in each preset time period, wherein the visualization layout at least comprises a display area of the target sentiment category and a form of the visualization graph.

13. The apparatus according to claim 8, wherein the data encoding means is configured to determine position information of the keyword in the visual graph according to time information of the keyword, and add a corresponding keyword at a corresponding position in the visual graph according to the position information, wherein the time information of the keyword is time information of public opinion data to which the keyword belongs.

14. The apparatus of claim 8, wherein the data encoding device is further configured to add a color to the visual graphics and the keyword, wherein the color is associated with an emotion category corresponding to the visual graphics.

15. A public opinion data visualization device, wherein the device comprises:

a processor; and

one or more machine-readable media having machine-readable instructions stored thereon that, when executed by the processor, cause the apparatus to perform the method of any of claims 1-7.

Images