CN109829033B - Data display method and terminal equipment - Google Patents
Data display method and terminal equipment Download PDFInfo
- Publication number
- CN109829033B CN109829033B CN201711182131.6A CN201711182131A CN109829033B CN 109829033 B CN109829033 B CN 109829033B CN 201711182131 A CN201711182131 A CN 201711182131A CN 109829033 B CN109829033 B CN 109829033B
- Authority
- CN
- China
- Prior art keywords
- emotion
- attribute
- data
- words
- polarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The application provides a data display method and terminal equipment, wherein the data display method comprises the following steps: determining a target object; acquiring multimedia data related to the target object from a plurality of data sources; determining feature words related to the target object from the multimedia data; and displaying the determined characteristic words. Compared with the existing mode that data are acquired only through one data source and displayed characteristic words are preset, the scheme provided by the embodiment has the advantages that the data source is random, the displayed result is random, large-scale data statistics can be conveniently carried out, the data statistics can be applied to public sentiment data, and effective integration and utilization of the public sentiment data can be realized.
Description
Technical Field
The application belongs to the technical field of internet, and particularly relates to a data display method and terminal equipment.
Background
With the rapid development of science and technology, especially the rapid development of information technology. People increasingly rely on the internet to obtain information, and more people also publish, through the internet, evaluations of the preference or role of a certain product, which may be referred to as public sentiment.
The public opinion data can be effectively integrated and utilized, so that more comprehensive and objective cognition can be provided for people. The conventional method for providing product cognition for users based on public opinion data is relatively single, generally only directional evaluation contents of users on one website are used as data sources, and the integration result is also based on preset characteristic words. For example, the evaluation of a restaurant is based on the user aggregating the contents of directional evaluations of the restaurant at a certain website, and the result of aggregation is also a preset frame, and the user's evaluation of the restaurant at the website, such as a score and contents of evaluations, is obtained by performing aggregation classification based on words in the frame.
Therefore, a fixed directional data source is needed, and the display result is relatively fixed, so that the public opinion data cannot be effectively analyzed and utilized.
No effective solution to this problem has been proposed so far.
Disclosure of Invention
The application aims to provide a data display method and terminal equipment, and public opinion data can be effectively integrated and utilized.
The application provides a data display method and terminal equipment, which are realized as follows:
a method of data presentation, comprising:
determining a target object;
acquiring multimedia data related to the target object from a plurality of data sources;
determining feature words related to the target object from the multimedia data;
and displaying the determined characteristic words.
A terminal device comprising a processor and a memory for storing processor-executable instructions, the instructions when executed by the processor performing the steps of:
determining a target object;
acquiring multimedia data related to the target object from a plurality of data sources;
determining feature words related to the target object from the multimedia data;
and displaying the determined characteristic words.
A computer readable storage medium having stored thereon computer instructions which, when executed, implement the steps of the above-described method.
According to the data display method, after the target object of query is determined, related multimedia data are obtained from a plurality of data sources, and the characteristic words of the target object are determined from the obtained multimedia data and displayed. Compared with the existing mode that data are acquired only through one data source and displayed characteristic words are preset, the scheme provided by the embodiment has the advantages that the data source is random, the displayed result is random, large-scale data statistics can be conveniently carried out, the data statistics can be applied to public sentiment data, and effective integration and utilization of the public sentiment data can be realized.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.
Fig. 1 is a schematic diagram illustrating an architecture of a public opinion data processing system according to an embodiment of the present application;
FIG. 2 is a schematic diagram of an embodiment of a clustering result presentation interface provided herein;
fig. 3 is a schematic diagram illustrating a public opinion data-based analysis process according to the present application;
FIG. 4 is a diagram illustrating one embodiment of a dependency syntax tree provided herein;
FIG. 5 is an exemplary diagram of a clustering result presentation interface provided herein;
fig. 6 is a schematic architecture diagram of an embodiment of a terminal device provided in the present application;
fig. 7 is a schematic block diagram of an embodiment of a data display apparatus provided in the present application.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In this example, it is considered that if description or evaluation data related to a specific product can be obtained from massive data of the internet, then attributes and affective information of multiple dimensions of the specific product are obtained from the data, and aggregation statistics is performed on the attributes and affective information to form evaluation cognitive information of multiple dimensions of the specific product, so that a user can more comprehensively perceive the specific product.
Based on this, in this example, a public opinion data processing system is provided, as shown in fig. 1, which may include: server 101, terminal 102. The server 101 may obtain multimedia data related to the target object from multiple data sources, determine a feature word related to the target object from the multimedia data, and push the feature word to the terminal 102 for display. It is also possible that these operations are all performed by the terminal 102, that is, data is obtained from a plurality of data sources, and the data is displayed in an integrated manner by the terminal 102. The specific mode is adopted, and the method is not limited in the application according to the actual needs.
In an embodiment, the server 101 may be a single server, a server cluster, a processor in a cloud, or the like, and which mode is specifically adopted may be selected according to actual needs.
In one embodiment, the terminal 102 may be a terminal device or software used by a user. Specifically, the terminal 102 may be a terminal device such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart watch, or other wearable devices. Of course, the terminal 102 may also be software that can run in the above-described terminal device. For example: and the mobile phone is applied to application software such as a Taobao, a Paobao or a browser.
In one embodiment, the multimedia data may be text data, voice data, video data, or the like. For example, if the acquired multimedia data is voice data, the text carried therein can be recognized by voice recognition, and if the acquired multimedia data is video data, the text carried therein can be recognized by image semantic understanding. In implementation, which type or types of data are selected as the acquired multimedia data may be selected according to actual needs, which is not limited in the present application.
The data acquisition and aggregation are performed in the server 101, but it is needless to say that the terminal 102 may perform the data acquisition and aggregation operations as follows.
Specifically, the data display method may include the following steps:
s1: determining a target object;
in one embodiment, the keyword input by the user may be received and the keyword input by the user may be used as the target object, or a keyword list may be set in advance, and the user clicks and uses the keyword clicked by the user as the target object. It is of course also possible that the user directly selects the keyword that was input last time as the target object and so on,
it should be noted, however, that the above listed method for determining a target object is only an exemplary description, and other methods for determining a target object may be adopted in practical implementation, which is not limited in this application.
Taking the car as an example, the user may input car, i.e., car as a keyword, and accordingly the determination target object is car.
S2: acquiring multimedia data related to the target object from a plurality of data sources;
after the target object "car a" is determined, data relating to the target object may be obtained from the internet. Specifically, when data is acquired, the data may be crawled from a plurality of website platforms. For example, the data may be obtained from a news website only, or the data related to the target object may be obtained from a news website, a forum, a post, an e-commerce platform, a microblog, or the like. That is, data may be acquired from multiple data sources, thereby making the acquired data about "car a" more comprehensive.
When determining the data source, it may be determined which data source is the target data source according to actual requirements. For example, if it is necessary to see the entire operation of the car, and a user needs to know a comprehensive understanding about news reports, forums, posts, and e-commerce websites on the internet, it is necessary to crawl data from multiple network platforms to obtain a more comprehensive statistical result about the car.
That is, the data acquired in this example is acquired from a plurality of data sources, such as news data, data of a post, and the like, which are all nondirectional, and are completely different from the conventional review website in which only the data of its own website is aggregated and displayed, and the data of the conventional review website is generally directional, and is data that is specifically reviewed for the subject.
S3: determining feature words related to the target object from the multimedia data;
the above feature words may include, but are not limited to, at least one of: attribute words, emotion words, attribute categories.
Specifically, after massive data is acquired from a plurality of data sources, sentences containing emotion expressions can be extracted from the massive data, so that the complexity of subsequent processing can be reduced, and the time for data processing can be saved. Of course, this is only an optional step, and in actual implementation, the data may not be preprocessed first, but all the data are directly considered as sentences containing emotional expressions, and the sentences are directly sent to the subsequent processing process. The specific method can be selected according to the processing capability and the actual requirement of the system, and the application is not limited to this.
For example: the target object is car, car object a is a car from country XX, and car object a is particularly comfortable, the first sentence is not an emotion sentence, and the second sentence is an emotion sentence. At the time of data preprocessing, only "car a is particularly comfortable" is extracted, and "car a object is a car from country XX" is discarded.
In the process of preprocessing data and extracting sentences containing emotional expressions, emotional sentences can be identified through a fasttext text classification algorithm. The fasttext text classification algorithm is a public text classification algorithm, multi-dimensional features are extracted from sentences, classification functions are learned, and the classification of the sentences is predicted.
However, it should be noted that the fasttext text classification algorithm mentioned above is only an exemplary description, and other text classification algorithms may be used to identify and obtain the emotion sentences in practical implementation, and the application is not limited thereto.
After the emotion sentences are determined, attribute-emotion pairs can be extracted from the acquired emotion sentences, and attribute types to which the attribute-emotion pairs belong can be determined.
For example, the emotional sentence is: after 500 yuan is broken through in the 14 th-day disk in 8 months, the small A car does not stand stably at the stock price, the oscillation descending situation appears for a few days later, and as the 8 th-18 days later, the small A car collects 489.65 yuan per stock, the attribute word can be extracted from the emotion sentence to be the stock price, the corresponding emotion descriptor is the oscillation descending, and then the attribute-emotion pair extracted based on the emotion sentence is the stock price-oscillation descending.
Further, the attribute category "market value" to which the attribute-emotion pair belongs may be determined.
Specifically, in the implementation, the attribute type may be preset according to an attribute system in which the target object is more core. For example, for "car a" the following attribute categories "comfort", "appearance", "historical culture", "market value", "strategic decision" and the like may be included, but are not limited to. These can be used as attribute categories to which attribute-emotion word pairs belong.
In one embodiment, the attribute category to which the attribute belongs may be determined based on the entire attribute-emotion pair, or may be determined based on only the attribute words in the attribute-emotion pair.
The attribute words or attribute-emotion pair sets corresponding to the attribute categories can be mined in advance based on the characteristics of the attribute categories. In this way, in the process of determining the attribute category, matching can be performed on the set corresponding to each attribute category based on the attribute words or the attribute-emotion pairs, and the attribute category with the highest similarity is selected as the determined attribute category.
However, it should be noted that the above listed manner for determining the attribute category is only an exemplary description, and other manners may be adopted in practical implementation, and the present application does not limit this.
Further, the sentiment polarity of an attribute or attribute-sentiment pair may be determined. I.e. whether it is positive, neutral or negative.
For example: the "concussion downlink" is negative and the "comfort" is positive.
For each attribute-emotion pair, a text sentence where the attribute-emotion pair is located can be extracted first, and then the emotion polarity of the attribute can be judged according to the emotion polarity of the text sentence and the polarity of the emotion words in the attribute-emotion pair in a combined manner, or the polarity of the attribute-emotion pair can be judged according to the corresponding polarity of the attribute-emotion pair. Which determination result is specifically obtained can be selected according to actual needs, which is not limited in the present application.
The emotion polarity of the words can be judged through a fasttext text classification algorithm, and the emotion polarity of the emotion words can be judged through a pre-established emotion dictionary, artificial features and the like.
However, it should be noted that the above-listed method for determining the emotion polarity is only an exemplary description, and other methods may be used for determining the emotion polarity when the emotion polarity is actually implemented, and the present application is not limited thereto.
After the massive network data are processed, a plurality of attribute emotion pairs can be obtained, and the types of the attribute emotion pairs and the emotion polarities of the attribute emotion pairs are determined. Based on the method, the attribute categories can be classified and integrated in the sentiment sentences, so that the user can conveniently check the public sentiment distribution corresponding to the target.
S4: and displaying the determined characteristic words.
In one embodiment, the presentation may be performed in a manner as shown in fig. 2, that is, the number of sentiment sentences in each attribute category of the target object, the attribute sentiment pair, the sentiment polarity of the attribute sentiment pair, and the like may be presented on a presentation interface.
Furthermore, the method can respond to a display request of a user for inquiring a corresponding emotion sentence of a certain attribute emotion pair, display the emotion sentence corresponding to the selected attribute emotion pair, synchronously display source websites of the emotion sentences, and highlight the attribute emotion pair related to the source websites, so that the user can effectively view the source of the attribute emotion pair and corresponding multimedia data information.
In an embodiment, considering that the emotional polarities of the attribute emotional pairs may also be displayed in the display interface, for example, the polarities may be indicated sequentially or in a designated ratio, or the polarities may be indicated by the color attribute, and which manner is specifically selected may be selected according to actual needs, which is not limited in this application.
For attribute emotion pairs, some attribute emotion pairs appear more frequently, indicating that the popularity is higher, and some attribute emotion pairs appear less frequently, and the relative popularity is lower. In order to characterize the degree of heat, the degree of heat may be identified by a number, a ratio, or the like, or may be identified by the distance from the core word (i.e., the target object). The specific selection of which mode can be selected according to actual needs, which is not limited in the present application.
In one embodiment, the result presentation may be a templatized presentation, i.e., the type of content to be presented by each display module in the display interface is predefined, and the operation items may be presented according to the predefined presentation.
In one embodiment, after the results are presented, a plurality of options may be provided, for example, a time period option may be provided and the user may select a time period. To select "within three days", the results from the aggregation of data for only three days can be displayed at this time. That is, a selected time period may be received; and presenting target characteristic words, wherein the target characteristic words are determined according to the multimedia data related to the target object, the release time of which on the data source is within the time period. By the method, the viewer can select the data in the time period to display according to actual needs, and the time period set for the system is not required to be limited, so that the viewing of the result is more convenient, and the flexibility is higher.
In one embodiment, the data presentation method can be applied to an evaluation website, a science popularization website, a product introduction website and the like, and the aggregation analysis processing and presentation of public opinion data can be realized through the processing method.
In this example, the presentation styles of the presentation results are completely different, and the presentation feature words are preset in the prior art, however, in this example, the presentation results are obtained by aggregating the multimedia data acquired in real time, that is, the presented feature words are obtained by temporarily aggregating the data, but are not preset, that is, the presentation results are all derived from the data acquired at present and are random.
The data display method in the above example can also be applied to the scene of data statistics query. For example, the old of the car always hopes to know the recent market feedback situation of the car, and the data feedback situation of the car on the internet within a certain time can be simply and intuitively seen through the public opinion data based data presentation method. For example: user public praise, stock market changes, news ratings for them, and the like.
In the above example, after the target object of the query is determined, related multimedia data are obtained from a plurality of data sources, and the feature words of the target object are determined from the obtained multimedia data and displayed. Compared with the existing mode that data are acquired only through one data source and displayed characteristic words are preset, the scheme provided by the embodiment has the advantages that the data source is random, the displayed result is random, large-scale data statistics can be conveniently carried out, the data statistics can be applied to public sentiment data, and effective integration and utilization of the public sentiment data can be realized.
The foregoing data processing and classifying method is described below with reference to a specific scenario, but it should be noted that the specific scenario is only for better explaining the present application and does not constitute a non-limiting limitation to the present application.
In this example, the information extraction and the classification display of the "car a" are described, specifically, the information of the "car a" in the internet is processed by combining text classification, information extraction, and the like, the emotional tendency and the emotional viewpoint of the user in each dimension attribute of the car a are extracted, and the information is displayed in a way of aggregation statistics.
As shown in fig. 3, the following process may be included:
s1: capturing data on platforms such as web pages, forums, posts, microblogs and the like;
s2: combining a text classification algorithm, and extracting emotional sentences with user emotional expressions from a large amount of drilled unstructured multimedia data;
considering that a large amount of texts unrelated to the target object (in this case, car of car a) exist in the web page text, if candidate sentences containing emotional expressions of the user can be mined, the workload of the subsequent extraction process can be effectively reduced, and the accuracy of attribute extraction can be provided. In order to achieve the purpose, a large amount of texts in the webpage texts can be extracted, and emotional sentences which are related to the target object and have user emotional expressions are obtained. For example, a recognition classification model can be learned from a manually labeled data set by, but not limited to, a fasttext text classification algorithm to determine whether an input text sentence is an emotional sentence with emotional expressions of a user related to a target object.
In one embodiment, the emotion sentences extracted with the emotion expressions of the users can be combined with subsequent information extraction to form a unified extraction model. For example, the sentiment score of a sentence can be determined by LSTM (Long Short-Term Memory network), and then the sentiment score is added to a CRF (Conditional Random Field algorithm) attribute extraction model as a feature of one dimension.
S3: performing attribute-level emotion analysis on the extracted emotion sentences with user emotion expressions, which may include:
a) Extracting attribute words, for example, extracting attribute words in a sentence through a sequence tagging algorithm;
b) Extracting the sentiment words, for example, the sentiment words in the sentence can be extracted according to the syntactic rules and the sequence tagging algorithm;
specifically, the extraction of the attribute words and the emotion words can be performed by combining the context characteristics, the part-of-speech characteristics and the dependency syntactic characteristics of the words.
The dependency syntax is used for parsing the sentence into a dependency syntax tree, and describing the dependency relationship among the words. That is, syntactic collocation relationships between words are indicated, which are semantically related. For example: the dependency syntax tree for the sentence "meeting announced the first list of senior universities" may be as shown in fig. 4. As can be seen from fig. 4, the word "announce" dominates "the conference", "has" and "has been listed", and thus these words are used as collocations of "announce".
Specifically, a CRF algorithm is used for extracting a candidate attribute word set in an emotion sentence, the CRF algorithm is used for extracting the candidate emotion word set of each attribute word, and information such as position, syntax and the like is fused in the process; then, the artificial rule can be extracted by combining the dependency syntax information, and the attribute-emotion word pair is extracted. And finally, performing cross verification on the attribute-emotion pair extracted based on the CRF algorithm and the attribute-emotion pair extracted based on the dependency syntax so as to finally determine the attribute-emotion pair.
The above listed manner of extracting attribute-emotion pairs is only an exemplary description, and other manners of extracting attribute-emotion pairs may be used in practical implementation, for example, other sequence labeling algorithms may be used, such as: hidden markov, maximum entropy markov models, etc. Meanwhile, the dependency syntax features used in the extraction part may also be presented in different forms, for example: the dependency syntax tree is first used to generate a vector representation of the dependency tree using a recurrent neural network as the CRF model input, and so on. The specific implementation form can be selected according to actual needs, and the application does not limit the implementation form.
C) Normalizing the attribute category, and mapping the attribute words to a certain attribute category in a preset attribute category set by combining an automatic mining dictionary and a similarity calculation model.
S3: attribute category normalization;
it is contemplated that the attribute categories will generally be different for different industries. Therefore, according to the difference of industry categories or the difference of products, a corresponding attribute category set can be set for each industry category in advance. For example, taking the liquor data as an example, the following attribute categories can be set but are not limited to: "comfort", "appearance", "historical culture", "market value", "strategic decision" etc.
In the attribute class normalization process, the extracted attribute-emotion pairs can be normalized to a specific attribute class. For example, a batch of candidate attribute-emotion pair sets can be automatically mined according to the characteristics of the attribute dimensions, then the average similarity between the mined attribute-emotion pairs and the word pair set in each attribute category is calculated, and the attribute category with the highest similarity is taken as the attribute category to which the attribute categories are normalized.
D) By the text classification algorithm, the emotion polarity of the attribute is judged (for example: positive, neutral, negative);
specifically, for each attribute word, a text sentence where the attribute is located may be extracted, and then the emotion polarity of the attribute is jointly determined by the emotion polarity of the text sentence and the polarity of the emotion word corresponding to the attribute word. The emotion polarity of the sentence can be judged through a fasttext text classification algorithm, and the emotion polarity of the emotion words can be judged through a pre-established emotion dictionary and artificial features (such as existence of negative words).
S4: for a batch structured attribute emotion analysis for a target object (a product or an enterprise) within a period of time, attribute emotion analysis results can be obtained, and aggregation statistics can be performed on the results. Specifically, the visualization display can be performed according to different heat degrees, emotional degrees, timeliness and the like.
Fig. 5 is a schematic presentation diagram showing product opinions based on statistical results, and as shown in fig. 5, part 1 supports a user to select opinions in different time ranges from a time dimension; section 2 shows different attribute systems for different types of consumer goods, and key attributes of the automotive industry are shown in fig. 5; the 3 rd part, namely the display of the extraction result of the attribute word emotion words, can display N attribute-emotion word pairs with higher heat for each attribute category, and respectively display the emotion polarity according to different colors, and can represent the statistical heat of the attribute through the distance between each word and the core word; section 4 supports user selection of different keywords; and 5, displaying related news original sentences and highlighting related text segments according to the keywords selected by the user.
However, it should be noted that this is only an exemplary presentation interface, and other aggregate presentation forms may be adopted when the actual implementation is performed, for example: the emotion words can be displayed according to the heat degree of the emotion words, the emotion degree of the attributes and the like, the specific display mode can be selected according to actual needs, and the emotion words are not limited in the application.
In the embodiment, by filtering out texts which do not contain emotion information in advance, attribute emotion pair extraction and cross validation are performed based on the filtered problems, so that the performance of attribute extraction is improved, and the emotion polarity corresponding to the attribute emotion pair is determined. For example, the extracted result covers a wider range, and may include public opinion information at the company level, such as: "historical culture", "market value", "high-level leader" of a company, etc. For the display of results, a more structured display scheme is provided, the display interface is more friendly, and the content is richer.
The embodiments of the data processing and displaying method provided by the present application can be executed in a server, a terminal or a similar computing device. Taking the operation on the terminal device as an example, fig. 6 is a hardware structure block diagram of the terminal device of the data display method according to the embodiment of the present invention. As shown in fig. 6, terminal device 10 may include one or more (only one shown) processors 102 (processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission module 106 for communication functions. It will be understood by those skilled in the art that the structure shown in fig. 6 is only an illustration and is not intended to limit the structure of the electronic device. For example, the computer terminal 10 may also include more or fewer components than shown in FIG. 6, or have a different configuration than shown in FIG. 6.
The memory 104 may be used to store software programs and modules of application software, such as program instructions/modules corresponding to the data presentation method in the embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implements the data presentation method of the application program. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the computer terminal 10 over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission module 106 is used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission module 106 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.
Referring to fig. 7, in a software implementation, the data presentation apparatus is applied in a terminal device, and may include: the device comprises a first determining unit, an obtaining unit, a second determining unit and a display unit. Wherein:
a first determination unit configured to determine a target object;
an acquisition unit configured to acquire multimedia data related to the target object from a plurality of data sources;
the second determining unit is used for determining the characteristic words related to the target object from the multimedia data;
and the display unit is used for displaying the determined characteristic words.
In one embodiment, the feature words may include, but are not limited to, at least one of: attribute words, emotion words, attribute categories.
In an embodiment, the second determining unit may specifically aggregate a plurality of feature words from the multimedia data, and the frequency of occurrence of each feature word; correspondingly, the display unit can specifically display the determined characteristic words and the occurrence frequency of the characteristic words.
In one embodiment, the apparatus may further include: the first receiving unit is used for receiving a query request for the target characteristic words after the determined characteristic words are displayed; the presentation unit may specifically present, in response to the query request, the multimedia data related to the target feature word.
In one embodiment, the apparatus may further include: the second receiving unit is used for receiving the selected time period after the determined characteristic words are displayed; the presentation unit may specifically present a target feature word, where the target feature word is determined according to the multimedia data related to the target object whose publication time on the data source is within the time period.
In one embodiment, the determining unit may specifically extract an emotional sentence with emotional expression, which describes the target object, from the multimedia data; finding out a plurality of attribute emotion word pairs from the emotion sentences; determining attribute types and emotion polarities to which each attribute emotion word pair belongs in the attribute emotion word pairs; performing aggregation classification according to the attribute type and the emotion polarity to which each attribute emotion word pair belongs in the plurality of attribute emotion word pairs to obtain the number of emotion sentences and the emotion polarity corresponding to each attribute type; and taking the attribute type and the emotion polarity to which each attribute emotion word pair belongs, and the number of emotion sentences and the emotion polarity corresponding to each attribute type as the feature words.
In one embodiment, the determining unit may specifically determine the attribute category and the emotion polarity to which the attribute emotion word pair belongs according to one of the following manners:
1) Determining the emotion polarity of the attribute emotion word pair according to the emotion polarity of the emotion words in the attribute emotion word pair;
2) Determining the emotion polarity of the attribute emotion word pair according to the emotion polarity of the emotion sentence where the attribute emotion word pair is located;
3) And determining the emotion polarity of the attribute emotion word pair according to the emotion polarity of the emotion words in the attribute emotion word pair and the emotion polarity of the emotion sentence in which the attribute emotion word pair is positioned.
Specifically, determining the emotion polarity of the attribute emotion word pair according to the emotion polarity of the emotion word in the attribute emotion word pair and the emotion polarity of the emotion sentence in which the attribute emotion word pair is located may include:
s1: determining the emotion polarity of the emotion words in the attribute emotion word pair;
s2: determining the emotion polarity of the attribute emotion word pair in the emotion sentence;
s3: and performing cross validation on the emotion polarity of the determined emotion words and the emotion polarity of the determined emotion sentences, and taking the cross validation result as the emotion polarity of the attribute emotion word pair.
The accuracy of emotion polarity identification is improved in a cross checking mode.
In the process of acquiring the multimedia data, the multimedia data related to the target object can be acquired from a plurality of target webpages, and then the emotional sentences with emotional expressions, which describe the target object, are extracted from the multimedia data.
In one embodiment, the attribute type may be an attribute classification set in advance for a characteristic of the target object.
In one embodiment, the attribute category may further include: and classifying the attributes of the enterprise to which the target object belongs.
In one embodiment, determining the attribute category to which the attribute emotion word pair belongs may include: comparing the similarity of the attribute emotion word pair with a pre-mined attribute emotion pair set aiming at each attribute category; and determining the attribute category corresponding to the highest similarity as the attribute category to which the attribute emotion word pair belongs.
In one embodiment, the sentimental polarity may include, but is not limited to, at least one of: positive emotion, neutral emotion, negative emotion.
After the target object of the query is determined, related multimedia data are obtained from a plurality of data sources, and the characteristic words of the target object are determined from the obtained multimedia data and displayed. Compared with the existing mode that the data is obtained only through one data source and the displayed characteristic words are preset, the scheme provided by the embodiment has random data sources and random displayed results, can facilitate large-scale data statistics, can be applied to the public opinion data, and can realize effective integrated utilization of the public opinion data.
Although the present application provides method steps as described in an embodiment or flowchart, additional or fewer steps may be included based on conventional or non-inventive efforts. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of sequences, and does not represent a unique order of performance. When an actual apparatus or client product executes, it may execute sequentially or in parallel (e.g., in the context of parallel processors or multi-threaded processing) according to the embodiments or methods shown in the figures.
The apparatuses or modules illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. For convenience of description, the above devices are described as being divided into various modules by functions, which are described separately. The functionality of the various modules may be implemented in the same one or more pieces of software and/or hardware when the application is implemented. Of course, a module that implements a certain function may be implemented by a plurality of sub-modules or sub-units in combination.
The methods, apparatus or modules described herein may be implemented in computer readable program code means for a controller implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, application Specific Integrated Circuits (ASICs), programmable logic controllers and embedded microcontrollers, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may therefore be considered as a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
Some of the modules in the apparatus described herein may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, classes, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
From the above description of the embodiments, it is clear to those skilled in the art that the present application can be implemented by software plus necessary hardware. Based on such understanding, the technical solutions of the present application may be embodied in the form of software products or in the implementation process of data migration, which essentially or partially contributes to the prior art. The computer software product may be stored in a storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to perform the methods described in the various embodiments or portions of the embodiments of the present application.
The embodiments in the present specification are described in a progressive manner, and the same or similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. All or portions of the present application are operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.
While the present application has been described with examples, those of ordinary skill in the art will appreciate that there are numerous variations and permutations of the present application without departing from the spirit of the application, and it is intended that the appended claims encompass such variations and permutations without departing from the spirit of the application.
Claims (13)
1. A method for displaying data, comprising:
determining a target object;
acquiring multimedia data related to the target object from a plurality of data sources;
determining feature words related to the target object from the multimedia data;
displaying the determined characteristic words;
determining feature words related to the target object from the multimedia data comprises the following steps:
extracting an emotional sentence with emotional expression for describing the target object from the multimedia data;
finding out a plurality of attribute emotion word pairs from the emotion sentences;
determining attribute types and emotion polarities to which each attribute emotion word pair belongs in the attribute emotion word pairs;
performing aggregation classification according to the attribute type and the emotion polarity to which each attribute emotion word pair belongs in the plurality of attribute emotion word pairs to obtain the number of emotion sentences and the emotion polarity corresponding to each attribute type;
and taking the attribute type and the emotion polarity to which each attribute emotion word pair belongs, and the number of emotion sentences and the emotion polarity corresponding to each attribute type as the feature words.
2. The method of claim 1, wherein the feature words comprise at least one of: attribute words, emotion words, attribute categories.
3. The method of claim 1, wherein determining feature words associated with the target object from the multimedia data comprises:
aggregating a plurality of feature words from the multimedia data and the frequency of occurrence of each feature word;
correspondingly, the determined feature words are displayed, and the method comprises the following steps:
and displaying the determined characteristic words and the occurrence frequency of the characteristic words.
4. The method of claim 1, wherein after presenting the determined feature words, the method further comprises:
receiving a query request for a target feature word;
and responding to the query request, and displaying the multimedia data related to the target characteristic words.
5. The method of claim 1, wherein after presenting the determined feature words, the method further comprises:
receiving a selected time period;
and presenting target characteristic words, wherein the target characteristic words are determined according to the multimedia data related to the target object, the release time of which on the data source is within the time period.
6. The method of claim 1, wherein determining the attribute category and emotion polarity to which the attribute emotion word pair belongs comprises at least one of:
determining the emotion polarity of the attribute emotion word pair according to the emotion polarity of the emotion words in the attribute emotion word pair;
determining the emotion polarity of the attribute emotion word pair according to the emotion polarity of the emotion sentence where the attribute emotion word pair is located;
and determining the emotion polarity of the attribute emotion word pair according to the emotion polarity of the emotion words in the attribute emotion word pair and the emotion polarity of the emotion sentence in which the attribute emotion word pair is positioned.
7. The method of claim 6, wherein determining the emotion polarity of the attribute emotion word pair according to the emotion polarity of the emotion word in the attribute emotion word pair and the emotion polarity of the emotion sentence in which the attribute emotion word pair is located comprises:
determining the emotion polarity of the emotion words in the attribute emotion word pair;
determining the emotion polarity of the attribute emotion word pair in the emotion sentence;
and performing cross validation on the emotion polarity of the determined emotion words and the emotion polarity of the determined emotion sentences, and taking the cross validation result as the emotion polarity of the attribute emotion word pair.
8. The method of claim 1, wherein the attribute category is an attribute classification preset for a characteristic of the target object.
9. The method of claim 8, wherein the attribute categories further comprise: and classifying the attributes of the enterprise to which the target object belongs.
10. The method of claim 1, wherein determining the attribute category to which the attribute emotion word pair belongs comprises:
comparing the similarity of the attribute emotion word pair with a pre-mined attribute emotion pair set aiming at each attribute category;
and determining the attribute category corresponding to the highest similarity as the attribute category to which the attribute emotion word pair belongs.
11. The method of claim 1, wherein the emotion polarities include at least one of: positive emotion, neutral emotion, negative emotion.
12. A terminal device comprising a processor and a memory for storing processor-executable instructions which, when executed by the processor, implement the steps of the method of any one of claims 1 to 11.
13. A computer readable storage medium having stored thereon computer instructions which, when executed, implement the steps of the method of any one of claims 1 to 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711182131.6A CN109829033B (en) | 2017-11-23 | 2017-11-23 | Data display method and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711182131.6A CN109829033B (en) | 2017-11-23 | 2017-11-23 | Data display method and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109829033A CN109829033A (en) | 2019-05-31 |
| CN109829033B true CN109829033B (en) | 2023-04-18 |
Family
ID=66858542
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711182131.6A Active CN109829033B (en) | 2017-11-23 | 2017-11-23 | Data display method and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109829033B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112580358A (en) * | 2019-09-30 | 2021-03-30 | 北京国双科技有限公司 | Text information extraction method, device, storage medium and equipment |
| CN111581479B (en) * | 2020-05-11 | 2021-04-13 | 北京精准沟通传媒科技股份有限公司 | One-stop data processing method and device, storage medium and electronic equipment |
| CN112052375A (en) * | 2020-09-30 | 2020-12-08 | 北京百度网讯科技有限公司 | Public opinion acquisition and word viscosity model training method and device, server and medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102033910A (en) * | 2010-11-19 | 2011-04-27 | 福建富士通信息软件有限公司 | Enterprise search engine technology based on multiple data resources |
| CN102073701A (en) * | 2010-12-30 | 2011-05-25 | 浪潮集团山东通用软件有限公司 | Semantic definition-based multi-data source data querying method |
| CN103678564A (en) * | 2013-12-09 | 2014-03-26 | 国家计算机网络与信息安全管理中心 | Internet product research system based on data mining |
| CN105354183A (en) * | 2015-10-19 | 2016-02-24 | Tcl集团股份有限公司 | Analytic method, apparatus and system for internet comments of household electrical appliance products |
| CN105574176A (en) * | 2015-12-21 | 2016-05-11 | 北京奇虎科技有限公司 | Hot word recommending method and device with combination of multiple data sources |
| CN105824898A (en) * | 2016-03-14 | 2016-08-03 | 苏州大学 | Label extracting method and device for network comments |
| CN106599065A (en) * | 2016-11-16 | 2017-04-26 | 北京化工大学 | Food safety online public opinion early warning system based on Storm distributed framework |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013013093A1 (en) * | 2011-07-20 | 2013-01-24 | Ness Computing, Inc. | Method and apparatus for quickly evaluating entities |
| KR20140078312A (en) * | 2012-12-17 | 2014-06-25 | 한국전자통신연구원 | Apparatus and system for providing sentimet analysis results based on text and method thereof |
-
2017
- 2017-11-23 CN CN201711182131.6A patent/CN109829033B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102033910A (en) * | 2010-11-19 | 2011-04-27 | 福建富士通信息软件有限公司 | Enterprise search engine technology based on multiple data resources |
| CN102073701A (en) * | 2010-12-30 | 2011-05-25 | 浪潮集团山东通用软件有限公司 | Semantic definition-based multi-data source data querying method |
| CN103678564A (en) * | 2013-12-09 | 2014-03-26 | 国家计算机网络与信息安全管理中心 | Internet product research system based on data mining |
| CN105354183A (en) * | 2015-10-19 | 2016-02-24 | Tcl集团股份有限公司 | Analytic method, apparatus and system for internet comments of household electrical appliance products |
| CN105574176A (en) * | 2015-12-21 | 2016-05-11 | 北京奇虎科技有限公司 | Hot word recommending method and device with combination of multiple data sources |
| CN105824898A (en) * | 2016-03-14 | 2016-08-03 | 苏州大学 | Label extracting method and device for network comments |
| CN106599065A (en) * | 2016-11-16 | 2017-04-26 | 北京化工大学 | Food safety online public opinion early warning system based on Storm distributed framework |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109829033A (en) | 2019-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108021929B (en) | Big data-based mobile terminal e-commerce user portrait establishing and analyzing method and system | |
| CN108153901B (en) | Knowledge graph-based information pushing method and device | |
| EP3244312B1 (en) | A personal digital assistant | |
| CN108885623B (en) | Semantic analysis system and method based on knowledge graph | |
| US9830386B2 (en) | Determining trending topics in social media | |
| CN107832338B (en) | Method and system for recognizing core product words | |
| CN108614832B (en) | Method and device for realizing user personalized commodity search | |
| CN110020162B (en) | User identification method and device | |
| CN111339406B (en) | Personalized recommendation method, device, equipment and storage medium | |
| CN109829033B (en) | Data display method and terminal equipment | |
| CN112733042A (en) | Recommendation information generation method, related device and computer program product | |
| US10901992B2 (en) | System and method for efficiently handling queries | |
| US11640420B2 (en) | System and method for automatic summarization of content with event based analysis | |
| CN112818111B (en) | Document recommendation method, device, electronic equipment and medium | |
| CN111339295A (en) | Method, apparatus, electronic device and computer readable medium for presenting information | |
| CN112085568A (en) | Commodity and rich media aggregation display method and device, electronic device and medium | |
| CN114880498B (en) | Event information display method and device, equipment and medium | |
| CN114036397B (en) | Data recommendation method, device, electronic equipment and medium | |
| CN113365138B (en) | Content display method and device, electronic equipment and storage medium | |
| KR102322212B1 (en) | Apparatus and method for recommending learning contents | |
| CN110990710B (en) | Resource recommendation method and device | |
| CN110532454B (en) | Search term recommendation method and device | |
| US20150186539A1 (en) | System and methods thereof for generation of widgets based on entities | |
| CN111782776A (en) | Method and device for realizing intention identification through slot filling | |
| CN111339124B (en) | Method, apparatus, electronic device and computer readable medium for displaying data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |