CN108170685B - Text emotion analysis method and device and computer readable storage medium - Google Patents
Text emotion analysis method and device and computer readable storage medium Download PDFInfo
- Publication number
- CN108170685B CN108170685B CN201810082070.4A CN201810082070A CN108170685B CN 108170685 B CN108170685 B CN 108170685B CN 201810082070 A CN201810082070 A CN 201810082070A CN 108170685 B CN108170685 B CN 108170685B
- Authority
- CN
- China
- Prior art keywords
- emotion
- text
- association rule
- polarity
- dimension
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/355—Class or cluster creation or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure provides a text emotion analysis method and device and a computer readable storage medium, and relates to the technical field of text mining. The text emotion analysis method comprises the following steps: selecting a matched emotion association rule according to the field of the text, wherein the emotion association rule comprises identifications of various emotion dimensions, regular expressions and emotion polarities; segmenting the text into short texts; and analyzing the short text by using an emotion association rule to acquire the emotion dimensionality and the emotion polarity appearing in the text. By the method, short texts segmented by the texts can be processed by adopting emotion association rules matched with the fields of the texts to obtain emotion dimensions and emotion polarities appearing in the texts, so that classification and angle of user evaluation and attitude of the user can be automatically determined, and utilization rate and analysis accuracy of evaluation contents are improved.
Description
Technical Field
The disclosure relates to the technical field of text mining, and in particular to a text emotion analysis method, a text emotion analysis device and a computer-readable storage medium.
Background
In recent years, the internet technology has been rapidly developed, a large amount of webpage data is generated, and subjective evaluations of users on various products can be obtained from the internet. The comment texts contain rich user emotion colors, the emotions in the comments are analyzed, the specified dimension bad comments in the comments are extracted, and merchants or management units can be helped to find problems existing in products in time and make targeted improvement. Such as: in the comments of a user on a hotel, the comment of the type of 'poor sanitation' or 'poor service attitude' often appears, and the hotel can quickly locate the problem and conveniently make targeted correction.
In the related art, the user is required to fill in the dimension classification of the evaluation while filling in the comment, such as environment evaluation, price evaluation, service attitude evaluation, product quality evaluation, and the like. However, such operations are cumbersome and not highly user-friendly, which results in that some users can randomly select the product when filling in, and the comment content and the classification are not matched, which results in a very large error when determining the public praise and the existing problems of the product according to the classification filled in by the users.
Disclosure of Invention
An object of the present disclosure is to improve the utilization rate of evaluation contents and the accuracy of analysis.
According to one aspect of the disclosure, a text emotion analysis method is provided, which includes: selecting a matched emotion association rule according to the field of the text, wherein the emotion association rule comprises identifications of various emotion dimensions, regular expressions and emotion polarities; segmenting the text into short texts; and analyzing the short text by using an emotion association rule to acquire the emotion dimensionality and the emotion polarity appearing in the text.
Optionally, the emotion association rule further includes an emotion degree keyword and an emotion degree; the text emotion analysis method further comprises the following steps: analyzing the short text by using an emotion association rule to acquire the emotion degree of emotion dimensionality appearing in the short text; and counting the emotional degree of the same emotional dimension in the text.
Optionally, negative words are also included in the emotion association rule; and determining that the emotion polarity of the short text is opposite to the emotion polarity of the matched emotion dimension in the emotion association rule according to the negative word.
Optionally, the regular expression includes a dimension keyword collocation and a stop word collocation; determining emotion dimensions matched with the short text according to the dimension keyword collocation; and eliminating emotion dimensions matched with the short text according to the stop word collocation.
Optionally, the method further comprises: determining a plurality of emotion dimensions of the domain matched with the evaluation according to the domain scene; an emotion association rule is generated for each emotion dimension.
Optionally, the generating of the emotion association rule for each emotion dimension comprises: generating a regular expression according to the key words, the key word combinations and the stop word combinations of the emotion dimensionalities; determining the emotion polarity of the regular expression; and generating an emotion association rule according to the regular expression, the emotion polarity and the negative words of the emotion dimensionalities.
By the method, short texts segmented by the texts can be processed by adopting emotion association rules matched with the fields of the texts to obtain emotion dimensions and emotion polarities appearing in the texts, so that classification and angle of user evaluation and attitude of the user can be automatically determined, and utilization rate and analysis accuracy of evaluation contents are improved.
According to another aspect of the present disclosure, a text emotion analyzing apparatus is provided, including: the rule selection unit is configured to select a matched emotion association rule according to the field of the text, wherein the emotion association rule comprises a plurality of emotion dimensionality identifications, regular expressions and emotion polarities; a text segmentation unit configured to segment a text into short texts; and the text analysis unit is configured to analyze the short text by utilizing the emotion association rule and acquire the emotion dimensionality and the emotion polarity appearing in the text.
Optionally, the emotion association rule further includes an emotion degree keyword and an emotion degree; the text analysis unit is further configured to: analyzing the short text by using an emotion association rule to acquire the emotion degree of emotion dimensionality appearing in the short text; and counting the emotional degree of the same emotional dimension in the text.
Optionally, negative words are also included in the emotion association rule; the text analysis unit is further configured to: and determining that the emotion polarity of the short text is opposite to the emotion polarity of the matched emotion dimension in the emotion association rule according to the negative word.
Optionally, the regular expression includes a dimension keyword collocation and a stop word collocation; the text analysis unit is further configured to: determining emotion dimensions matched with the short text according to the dimension keyword collocation; and eliminating emotion dimensions matched with the short text according to the stop word collocation.
Optionally, the method further comprises: an emotion rule generation unit configured to determine a plurality of emotion dimensions of a domain whose evaluation matches, from the domain scene; an emotion association rule is generated for each emotion dimension.
Optionally, the generating of the emotion association rule for each emotion dimension comprises: generating a regular expression according to the key words, the key word combinations and the stop word combinations of the emotion dimensionalities; determining the emotion polarity of the regular expression; and generating an emotion association rule according to the regular expression, the emotion polarity and the negative words of the emotion dimensionalities.
According to still another aspect of the present disclosure, a text emotion analyzing apparatus is provided, including: a memory; and a processor coupled to the memory, the processor configured to perform any of the above text sentiment analysis methods based on the instructions stored in the memory.
The device can process the short text segmented by the text by adopting the emotion association rule matched with the field of the text to obtain the emotion dimensionality and the emotion polarity appearing in the text, thereby automatically determining the classification and angle of user evaluation and the attitude of the user, and improving the utilization rate of the evaluation content and the accuracy of analysis.
According to yet another aspect of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored, which instructions, when executed by a processor, implement the steps of any of the above text emotion analysis methods.
By executing the program instructions on the computer-readable storage medium, the short texts segmented by the texts can be processed by adopting the emotion association rules matched with the fields of the texts, and the emotion dimensionality and the emotion polarity appearing in the texts are obtained, so that the classification and angle of user evaluation and the attitude of the user are automatically determined, and the utilization rate and the analysis accuracy of the evaluation content are improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and not to limit the disclosure. In the drawings:
FIG. 1 is a flow chart of one embodiment of a text sentiment analysis method of the present disclosure.
FIG. 2 is a flowchart of another embodiment of a text sentiment analysis method of the present disclosure.
FIG. 3 is a flow chart of yet another embodiment of a text sentiment analysis method of the present disclosure.
FIG. 4 is a schematic diagram of an embodiment of a text emotion analysis apparatus according to the present disclosure.
FIG. 5 is a schematic diagram of another embodiment of a text emotion analysis apparatus according to the present disclosure.
FIG. 6 is a schematic diagram of a text emotion analysis apparatus according to still another embodiment of the present disclosure.
Detailed Description
The technical solution of the present disclosure is further described in detail by the accompanying drawings and examples.
A flow diagram of one embodiment of a text sentiment analysis method of the present disclosure is shown in fig. 1.
In step 101, selecting a matching emotion association rule according to the field of the text, wherein the emotion association rule comprises a plurality of emotion dimension identifications, regular expressions and emotion polarities.
In one embodiment, the related field may be determined according to the position of the extracted text, for example, if the text is extracted from the evaluation after the user purchases the goods, the field may be determined according to the category of the goods purchased by the user, for example, if the user purchases a scenic spot ticket, the field is a scenic spot; if the user purchases the food, the field is food; if the user purchases an electric appliance, the field is the electric appliance and the like. In another embodiment, the related fields can be determined according to keywords in the text, such as "hotel" in the text "environment of the hotel is general", and the text is determined to belong to the catering or accommodation field. The specific domain classification can be divided as desired.
In step 102, the text is segmented into short texts. In one embodiment, the text may be classified with punctuation as a separator. In one embodiment, words such as "and", etc. may also be provided as delimiters.
In step 103, the short text is analyzed by using the emotion association rule to obtain the emotion dimensionality and the emotion polarity appearing in the text. In one embodiment, the emotion dimensions and emotion polarities in each short text can be acquired first, and then the full text is summarized and counted to obtain the user evaluation analysis result. In another embodiment, the characters which embody the emotion of the user can be displayed in a prominent font and form, so that the characters are convenient for the user to view. For example, for the text "scenery so bad that the toilet is clean, but not worth 100 tickets", the background stores after recognition are: the scene is poor, the toilet is clean, but the price is not 100 entrance tickets, and the corresponding characters are displayed on the page in bold or red.
By the method, short texts segmented by the texts can be processed by adopting emotion association rules matched with the fields of the texts to obtain emotion dimensions and emotion polarities appearing in the texts, so that classification and angle of user evaluation and attitude of the user can be automatically determined, and utilization rate and analysis accuracy of evaluation contents are improved. The analysis result can be used as the evaluation of the merchant for the user to refer on one hand, and can also be used as the evaluation of the industry for the merchant on the other hand, and in addition, the merchant can also know the user evaluation to improve the self condition.
The emotion dimension can be identified by a code or a character representing the emotion dimension, such as price, environment, cost performance, quality, taste, hygiene, speed, service and the like.
The regular expression may include dimension keyword collocation, that is, a combination of words commonly used by the user when the emotional dimension is embodied, for example, the combination of words commonly used when evaluating the product price includes a combination of "price", "ticket price", and the like, and a combination of "expensive", "cheap", "high", "low", and the like, which may be a close combination of adjacent words, and may also be other words spaced in the middle. In one embodiment, the regular expression also comprises stop word collocation, and the words such as "Guizhou" and "Guiyang" are not related to emotional evaluation, but should be used as stop words although the words are expensive.
The emotion polarity can be positive polarity or negative polarity, which respectively represents whether the text analysis result is positive evaluation or negative evaluation. Information such as good evaluation, poor evaluation rate and the like can be obtained through the statistics of the emotional polarity, and basic and visual evaluation is obtained.
In one embodiment, the emotion association rule can also comprise a negative word, such as 'No', and after the negative word is matched, the emotion polarity of the short text is determined to be opposite to the polarity of the regular expression matching.
In one embodiment, a flow diagram of another embodiment of a text sentiment analysis method of the present disclosure is shown in FIG. 2.
In step 201, a matching emotion association rule is selected according to the field of the text, wherein the emotion association rule includes a plurality of emotion dimension identifiers, regular expressions and emotion polarities. The emotion association rule can also comprise an emotion degree keyword and an emotion degree, such as very much, very much and the like which indicate strong emotion, and somewhat little and the like which indicate weak emotion.
In step 202, the text is segmented into short text. In one embodiment, the text may be classified with punctuation or predetermined other characters as delimiters.
In step 203, the short texts are analyzed by using the emotion association rule to obtain the emotion dimensionality and the emotion polarity appearing in each short text.
In step 204, the short texts are analyzed by using the emotion degree keywords and the emotion degrees in the emotion association rules, and the emotion degree of each short text is determined. Most emotion degree keywords are adverbs. In one embodiment, each emotion degree may be weighted, for example, the emotion degree keyword "very" and "natural" has a weight of 1, a weight of "very" is 0.8, a weight of "compare" is 0.4, a weight of "slight" is 0.1, and so on.
In step 205, the emotion degrees of the short texts with the same emotion dimensionality and the same emotion polarity are counted to obtain the emotion degree of the text, so as to realize the refinement degree of emotion analysis. In one embodiment, the degree can be presented to the user, so that more refined display is achieved, and user experience is improved. In one embodiment, the rating level may be refined, and the rating may be performed using "good rating", "medium rating", "bad rating", or the like.
By the method, the emotion of each dimension in the text can be finely analyzed and counted through degree analysis, the information quantity obtained by emotion analysis is increased, and the analysis effect and the user experience are improved.
In one embodiment, emotion association rules for various fields can be established according to actual conditions. A flow chart of yet another embodiment of a text sentiment analysis method of the present disclosure is shown in fig. 3.
In step 301, a plurality of sentiment dimensions of a domain matching the rating is determined from the domain scene. In one embodiment, as with food, its emotional dimensions may include taste, price, hygiene, aesthetics, environment, quality of service, etc.; for tickets, the emotional dimensions may include context, size, scenery, etc.; for hotels and other commodities, the emotion dimensionality can be selected according to actual conditions. In one embodiment, the sentiment dimension can be set from positive and negative angles, respectively, such as setting sentiment dimension "expensive" and "cheap" for price.
In step 302, an emotion association rule is generated for each emotion dimension. In one embodiment, a regular expression can be generated according to the keywords, the keyword combination and the stop word combination of the emotion dimensions, and then the emotion polarity of the regular expression is determined.
In one embodiment, the emotion polarities can be determined according to the actual situation of the dimension angle, and the emotion polarities of the same character and word in different dimensions are different, for example, in the price angle, the emotion polarity of the dimension of price "high" is negative polarity, the emotion polarity of the dimension of price "low" is positive polarity, but in the service level angle, the emotion polarity of the dimension of service quality "high" is positive polarity, and the emotion polarity of the dimension of service quality "low" is negative polarity.
In one embodiment, the sentiment polarity may be set with reference to the formed set of words, e.g., a "positive sentiment" word includes: love, admire, happiness, feeling the same person, curiosity, cheering, dreaminess, appreciation, and the like; the words of 'negative emotion' comprise grief and semi-suspicion, have the advantages of being kept away from sight, unsatisfied, not taste, repentance, disapproval and the like; the words of 'positive evaluation' include indispensable, local excellence, talent-high eight-fill, sinking the fish and dropping the goose, stimulating the human, listening, and exercising the child; the term "negative rating" such as: ugly, bitter, overproof, gorgeous but unrealistic, desolate and cool, turbid, light and abnormal, high price, no matter in cavities, etc.
And generating an emotion association rule according to the regular expression, the emotion polarity and the negative words of the emotion dimensionalities. In one embodiment, the emotion association rules may be as shown in Table 1:
TABLE 1 Emotion association rule storage example
| Dimension (d) of | Regular expression | Negative word | Emotional polarity |
| Price | Cost/performance/price ratio | Is not limited to | Is just |
Wherein,. is a wildcard character.
In one embodiment, the emotion degree keywords and the negative words can be supplemented for the emotion association rule based on the degree adverb and the negative word lexicon.
In step 303, the matching emotion association rule is selected based on the domain of the text. In one embodiment, the emotion association rule may be selected by manual selection or automatic matching.
In step 304, the text is segmented into short text. In one embodiment, the text may be classified with punctuation as a separator.
In step 305, the short text is analyzed by using the emotion association rule to obtain the emotion dimension and the emotion polarity appearing in the text. In one embodiment, the emotional degree may also be determined. In one embodiment, when a negative word appears during the matching of the positive emotion dimensions for the same angle, the rating is categorized as corresponding negative emotion dimensions. If the text shows that the price is not expensive, the dimension of the price angle is determined to be matched with the dimension of the price angle, but the negative word 'not' is matched, and the text is classified into the dimension of the price angle, namely 'cheap'.
In step 306, adjusting the regular expression according to the test result or the feedback of the user, and executing step 302; or the emotion dimension can be adjusted according to the test result or the feedback of the user, and step 301 is executed.
In the related text analysis technology, classified texts need to be used as a training set, a classifier is constructed after text features are extracted, the texts are processed by the classifier, a large-scale corpus is needed for constructing the classifier, the quantity and quality of the corpus directly influence the performance of the classifier, and the high-quality corpus is generally difficult to obtain and poor in flexibility in the field of refinement. Through the method of the embodiment, the emotion association rule can be generated in a targeted manner, adjustment is facilitated, on one hand, the requirement is met, meanwhile, the system is miniaturized, the requirement on the capacity of processing equipment is reduced, the processing efficiency is improved, on the other hand, the method is more flexible, and the analysis quality is improved.
A schematic diagram of one embodiment of a text sentiment analysis device of the present disclosure is shown in FIG. 4. Rule selection unit 401 can select a matching emotion association rule according to the field of the text, where the emotion association rule includes multiple emotion dimension identifiers, regular expressions, and emotion polarities. In one embodiment, the emotion association rule may be selected by manual selection or automatic matching. The text segmentation unit 402 is capable of segmenting text into short text. In one embodiment, the text may be classified with punctuation as a separator. In one embodiment, words such as "and", etc. may also be provided as delimiters. Text analysis section 403 can analyze the short text by using the emotion association rule to obtain the emotion dimension and emotion polarity appearing in the text. In one embodiment, the emotion dimensions and emotion polarities in each short text can be acquired first, and then the full text is summarized and counted to obtain user evaluation analysis. In another embodiment, the words that represent the emotion of the user can be highlighted for the user to view.
The device can process the short text segmented by the text by adopting the emotion association rule matched with the field of the text to obtain the emotion dimensionality and the emotion polarity appearing in the text, thereby automatically determining the classification and angle of user evaluation and the attitude of the user, and improving the utilization rate of the evaluation content and the accuracy of analysis.
In one embodiment, the emotion association rule may further include an emotion degree keyword and an emotion degree, such as very, extraordinary, extreme, and very, etc. indicating strong emotion, somewhat, and slightly, etc. indicating weak emotion. Text analysis section 403 can analyze the short text by using the emotion degree keyword and the emotion degree in the emotion association rule, and determine the emotion degree of each short text. Most emotion degree keywords are adverbs. And counting the emotion degrees of the short texts with the same emotion dimensionality and the same emotion polarity to obtain the emotion degree of the text and realize the refinement degree of emotion analysis. In one embodiment, the degree can be presented to the user, so that more refined display is achieved, and user experience is improved.
In one embodiment, the emotion association rule may further include a negative word, such as "no", and when the text analysis unit 403 matches the negative word, it is determined that the emotion polarity of the short text is opposite to the polarity of the regular expression matching. By the method, text analysis of different expression habits can be correctly analyzed, and missing judgment and misjudgment are reduced.
In one embodiment, the regular expression may further include stop word collocation, and when the text analysis unit 403 matches the stop word collocation, it is determined that the emotion assessment is not related, so as to reduce misjudgment.
In one embodiment, as shown in fig. 4, the text emotion analysis apparatus may further include an emotion rule generation unit 404, which is capable of determining a plurality of emotion dimensions of the domain matching the evaluation according to the domain scene, and further generating an emotion association rule for each emotion dimension. Emotion rule generation section 404 supplies the generated emotion association rule to rule selection section 401 and text analysis section 403.
The device can generate emotion association rules in a targeted manner, is convenient to adjust, meets the requirements, reduces the requirements on the capacity of processing equipment, has higher flexibility and improves the analysis quality.
Fig. 5 is a schematic structural diagram of an embodiment of an emotion analyzing apparatus according to the present disclosure. The text emotion analysis device includes a memory 501 and a processor 502. Wherein: the memory 501 may be a magnetic disk, flash memory, or any other non-volatile storage medium. The memory is used for storing the instructions in the corresponding embodiments of the text emotion analysis method above. The processor 502 is coupled to the memory 501 and may be implemented as one or more integrated circuits, such as a microprocessor or microcontroller. The processor 502 is configured to execute instructions stored in the memory, and can automatically determine the classification and angle of the user evaluation and the attitude of the user, and improve the utilization rate of the evaluation content and the accuracy of the analysis.
In one embodiment, as also shown in FIG. 6, text emotion analysis apparatus 600 includes a memory 601 and a processor 602. The processor 602 is coupled to the memory 601 by a BUS 603. The text emotion analysis apparatus 600 may be connected to an external storage device 605 through a storage interface 604 to call external data, and may be connected to a network or another computer system (not shown) through a network interface 606. And will not be described in detail herein.
In the embodiment, the data instructions are stored in the memory, and the instructions are processed by the processor, so that the classification and the angle of the user evaluation and the attitude of the user can be automatically determined, and the utilization rate of the evaluation content and the accuracy of analysis are improved.
In another embodiment, a computer readable storage medium has stored thereon computer program instructions which, when executed by a processor, implement the steps of the method in the corresponding embodiment of the text emotion analysis method. As will be appreciated by one skilled in the art, embodiments of the present disclosure may be provided as a method, apparatus, or computer program product. Accordingly, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Thus far, the present disclosure has been described in detail. Some details that are well known in the art have not been described in order to avoid obscuring the concepts of the present disclosure. It will be fully apparent to those skilled in the art from the foregoing description how to practice the presently disclosed embodiments.
The methods and apparatus of the present disclosure may be implemented in a number of ways. For example, the methods and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order for the steps of the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless specifically stated otherwise. Further, in some embodiments, the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
Finally, it should be noted that: the above examples are intended only to illustrate the technical solutions of the present disclosure and not to limit them; although the present disclosure has been described in detail with reference to preferred embodiments, those of ordinary skill in the art will understand that: modifications to the specific embodiments of the disclosure or equivalent substitutions for parts of the technical features may still be made; all such modifications are intended to be included within the scope of the claims of this disclosure without departing from the spirit thereof.
Claims (10)
1. A text emotion analysis method comprises the following steps:
determining a plurality of emotion dimensions of the domain matched with the evaluation according to the domain scene;
generating an emotion association rule for each emotion dimension, comprising: generating a regular expression according to the key words, the key word combinations and the stop word combinations of the emotion dimensionalities; determining the emotion polarity of the regular expression; generating the emotion association rule according to the regular expression, the emotion polarity and the negative words of the emotion dimensionality;
selecting matched emotion association rules according to the field of the text, wherein the selecting comprises the following steps: determining the field of the text according to the keywords of the text; the emotion association rule comprises identification of various emotion dimensions, a regular expression and emotion polarity, wherein the regular expression comprises dimension keyword collocation and stop word collocation;
dividing the text into short texts according to punctuation marks and characters serving as separators;
analyzing the short text by using the emotion association rule to acquire the emotion dimensionality and the emotion polarity appearing in the text.
2. The method of claim 1, wherein the emotion association rule further comprises an emotion degree keyword and an emotion degree;
further comprising:
analyzing the short text by using the emotion association rule to acquire the emotion degree of the emotion dimensionality appearing in the short text;
and counting the emotional degree of the same emotional dimension in the text.
3. The method of claim 1, wherein the emotion association rule further comprises a negative word;
and determining that the emotion polarity of the short text is opposite to the emotion polarity of the matched emotion dimension in the emotion association rule according to the negative word.
4. The method of claim 1, wherein,
determining emotion dimensions matched with the short text according to the dimension keyword collocation;
and eliminating the emotion dimensionality matched with the short text according to the stop word collocation.
5. A text emotion analysis apparatus comprising:
an emotion rule generation unit configured to determine a plurality of emotion dimensions of a domain whose evaluation matches, from the domain scene; generating an emotion association rule for each emotion dimension, comprising: generating a regular expression according to the key words, the key word combinations and the stop word combinations of the emotion dimensionalities; determining the emotion polarity of the regular expression; generating the emotion association rule according to the regular expression, the emotion polarity and the negative words of the emotion dimensionality;
the rule selection unit is configured to select matched emotion association rules according to the field of the text, and comprises the following steps: determining the field of the text according to the keywords of the text; the emotion association rule comprises identification of various emotion dimensions, a regular expression and emotion polarity, wherein the regular expression comprises dimension keyword collocation and stop word collocation;
a text segmentation unit configured to segment the text into short texts according to punctuation marks and characters as separators;
and the text analysis unit is configured to analyze the short text by utilizing the emotion association rule to acquire the emotion dimension and the emotion polarity appearing in the text.
6. The device of claim 5, wherein the emotion association rule further comprises an emotion degree keyword and an emotion degree;
the text analysis unit is further configured to:
analyzing the short text by using the emotion association rule to acquire the emotion degree of the emotion dimensionality appearing in the short text;
and counting the emotional degree of the same emotional dimension in the text.
7. The apparatus of claim 5, wherein the emotion association rule further comprises a negative word;
the text analysis unit is further configured to: and determining that the emotion polarity of the short text is opposite to the emotion polarity of the matched emotion dimension in the emotion association rule according to the negative word.
8. The apparatus of claim 5, wherein,
the text analysis unit is further configured to:
determining emotion dimensions matched with the short text according to the dimension keyword collocation;
and eliminating the emotion dimensionality matched with the short text according to the stop word collocation.
9. A computer processing device, comprising:
a memory; and
a processor coupled to the memory, the processor configured to perform the method of any of claims 1-4 based on instructions stored in the memory.
10. A computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the steps of the method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810082070.4A CN108170685B (en) | 2018-01-29 | 2018-01-29 | Text emotion analysis method and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810082070.4A CN108170685B (en) | 2018-01-29 | 2018-01-29 | Text emotion analysis method and device and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108170685A CN108170685A (en) | 2018-06-15 |
| CN108170685B true CN108170685B (en) | 2021-10-29 |
Family
ID=62515671
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810082070.4A Active CN108170685B (en) | 2018-01-29 | 2018-01-29 | Text emotion analysis method and device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108170685B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109522412B (en) * | 2018-11-14 | 2021-02-26 | 鼎富智能科技有限公司 | Text emotion analysis method, device and medium |
| CN110364185B (en) * | 2019-07-05 | 2023-09-29 | 平安科技（深圳）有限公司 | Emotion recognition method based on voice data, terminal equipment and medium |
| CN110377739B (en) * | 2019-07-19 | 2022-03-04 | 出门问问(苏州)信息科技有限公司 | Text emotion classification method, readable storage medium and electronic device |
| CN110888983B (en) * | 2019-11-26 | 2022-07-15 | 厦门市美亚柏科信息股份有限公司 | Positive and negative emotion analysis method, terminal equipment and storage medium |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8463595B1 (en) * | 2012-03-06 | 2013-06-11 | Reputation.Com, Inc. | Detailed sentiment analysis |
| CN103679462B (en) * | 2012-08-31 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of comment data treating method and apparatus, a kind of searching method and system |
| EP2994846A4 (en) * | 2013-05-09 | 2017-01-25 | Moodwire, Inc. | Hybrid human machine learning system and method |
| CN103399916A (en) * | 2013-07-31 | 2013-11-20 | 清华大学 | Internet comment and opinion mining method and system on basis of product features |
| CN104484336B (en) * | 2014-11-19 | 2017-12-19 | 湖州师范学院 | A kind of Chinese comment and analysis method and its system |
| CN106407236B (en) * | 2015-08-03 | 2019-07-23 | 北京众荟信息技术股份有限公司 | A kind of emotion tendency detection method towards comment data |
| 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 |
| CN106897439B (en) * | 2017-02-28 | 2020-04-14 | 百度在线网络技术（北京）有限公司 | Text emotion recognition method, device, server and storage medium |
| CN107122471B (en) * | 2017-05-02 | 2020-07-10 | 北京众荟信息技术股份有限公司 | Hotel characteristic comment extraction method |
-
2018
- 2018-01-29 CN CN201810082070.4A patent/CN108170685B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108170685A (en) | 2018-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108170685B (en) | Text emotion analysis method and device and computer readable storage medium | |
| CN108628833B (en) | Method and device for determining summary of original content and method and device for recommending original content | |
| CN104281699B (en) | Method and device is recommended in search | |
| CN107870896B (en) | Conversation analysis method and device | |
| KR101423544B1 (en) | Device and method for extracting semantic topics | |
| CN106919661B (en) | Emotion type identification method and related device | |
| US20140172415A1 (en) | Apparatus, system, and method of providing sentiment analysis result based on text | |
| Herzig et al. | Predicting customer satisfaction in customer support conversations in social media using affective features | |
| CN110287405B (en) | Emotion analysis method, emotion analysis device and storage medium | |
| JP6851894B2 (en) | Dialogue system, dialogue method and dialogue program | |
| US20160117954A1 (en) | System and method for automated teaching of languages based on frequency of syntactic models | |
| US11126783B2 (en) | Output apparatus and non-transitory computer readable medium | |
| CN109213860A (en) | Merge the text sentiment classification method and device of user information | |
| CN105955957B (en) | The determination method and device that aspect scores in a kind of businessman's general comment | |
| KR101319413B1 (en) | Summary Information Generating System and Method for Review of Product and Service | |
| CN111737961B (en) | Method and device for generating story, computer equipment and medium | |
| CN106471494A (en) | For analyzing sales promotion website to the program of the effect that user psychology state changes, apparatus and method | |
| CN107797981B (en) | Target text recognition method and device | |
| Chatterjee et al. | Verbal behaviors and persuasiveness in online multimedia content | |
| JP5942052B1 (en) | Data analysis system, data analysis method, and data analysis program | |
| KR20200064490A (en) | Server and method for automatically generating profile | |
| CN110222181B (en) | Python-based film evaluation emotion analysis method | |
| John et al. | Topic modeling to extract information from nutraceutical product reviews | |
| CN111651590A (en) | Data processing method and device, electronic equipment and storage medium | |
| Wang et al. | Word intuition agreement among Chinese speakers: a Mechanical Turk-based study |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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 |