CN110888983A - Positive and negative emotion analysis method, terminal device and storage medium - Google Patents
Positive and negative emotion analysis method, terminal device and storage medium Download PDFInfo
- Publication number
- CN110888983A CN110888983A CN201911171315.1A CN201911171315A CN110888983A CN 110888983 A CN110888983 A CN 110888983A CN 201911171315 A CN201911171315 A CN 201911171315A CN 110888983 A CN110888983 A CN 110888983A
- Authority
- CN
- China
- Prior art keywords
- emotion
- industry
- text data
- positive
- negative
- 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.)
- Granted
Links
Images
Classifications
-
- 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
- G06F16/353—Clustering; Classification into predefined classes
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/374—Thesaurus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a positive and negative emotion analysis method, terminal equipment and a storage medium, wherein the method comprises the following steps: s1: establishing and maintaining industry keyword rules and industry emotion dictionaries corresponding to different industries; s2: judging whether the text data to be analyzed contains keywords contained in the industry keyword rule, and if yes, entering S3; otherwise, go to S4; s3: judging all industries to which the text data belongs according to the industries to which the keywords belong and the industry keyword rules corresponding to the industries, and then calculating the emotion score of each industry according to the industry emotion dictionary corresponding to each industry so as to obtain the emotion positive and negative analysis result of the text data; s4: and obtaining the positive and negative emotion analysis result of the text data through the trained machine learning model. The invention adopts an emotion analysis method based on the fusion of industry emotion words and a machine learning model, divides and treats the network texts, and improves the analysis effect.
Description
Technical Field
The invention relates to the technical field of text analysis, in particular to a positive and negative emotion analysis method, terminal equipment and a storage medium.
Background
Explosive growth of network data places more and higher demands on the analysis of data. The text analysis and mining technology is widely applied at present, semantic contents of texts are extracted through corresponding technologies and methods, and then a series of operations such as classification, clustering, positive and negative emotion analysis and the like are performed on the texts, and the method is mainly used for the fields of commodity recommendation, public opinion analysis, text search and the like.
In public opinion analysis, public opinions in a network need to be sorted and analyzed under different topics, for example, positive and negative emotion analysis is performed on collected texts, positive and negative emotion trends of postscripts are automatically identified, enterprises and governments can better grasp development conditions of the network public opinions, and a foundation is provided for public opinion guidance later. Therefore, the emotion positive and negative analysis of the collected text is a relatively important link in public opinion analysis.
In the related technology of text classification, the current emotion analysis is mainly divided into two methods, namely dictionary-based and machine learning-based. The method based on the emotion dictionary mainly needs a reliable knowledge base such as the emotion dictionary, and the method based on the machine learning needs a large amount of labeled samples. The text data to be analyzed in the public opinion field is very complicated, and can be generally divided into general fields and industrial fields according to business requirements. In the general field, the high-quality general field emotion dictionaries are very few, and in addition, semantic information is not fully considered based on the rule-based algorithm characteristics of the emotion dictionaries, so that the effect of using the algorithm based on the emotion dictionaries in the general field is not ideal. The algorithm based on machine learning, especially deep learning, is emerging in recent years, semantic information can be well combined, and in addition, the general field has a plurality of open source samples or open source models with good quality, so that the emotion analysis of the general field can be well represented based on the deep learning algorithm. However, in the field of special industries, manual collection of high-quality emotion knowledge bases becomes possible, emotion analysis in the special industries has relatively little dependence on semantics, good effects can be obtained through the emotion knowledge bases, and collection of a large number of industry label samples is time-consuming and labor-consuming.
Disclosure of Invention
Based on the problems, the invention provides a positive and negative face emotion analysis method, terminal equipment and a storage medium, wherein an emotion analysis method based on the fusion of industry emotion words and a machine learning model is adopted, web texts are divided and treated, industry field data (such as finance and economics and the like) are analyzed through an algorithm based on an industry emotion dictionary, general field data are analyzed through an algorithm of the machine learning model, and the overall analysis effect is improved.
The specific scheme is as follows:
a positive and negative face emotion analysis method comprises the following steps:
s1: establishing and maintaining an industry keyword rule and an industry emotion dictionary corresponding to different industries, wherein the industry emotion dictionary comprises main words capable of expressing the type of the industry and emotion words capable of expressing the industry emotion;
s2: judging whether the text data to be analyzed contains keywords contained in the industry keyword rule, and if yes, entering S3; otherwise, go to S4;
s3: judging all industries to which the text data belongs according to the industries to which the keywords belong and the industry keyword rules corresponding to the industries, and then calculating the emotion score of each industry according to the industry emotion dictionary corresponding to each industry so as to obtain the emotion positive and negative analysis result of the text data;
s4: and obtaining the positive and negative emotion analysis result of the text data through the trained machine learning model.
Further, the method for determining all industries to which the text data belongs in step S3 is as follows: and calculating Boolean operation results of the industry keyword rules corresponding to each industry according to all keywords contained in the text data, and judging whether the industry belongs to the industry or not according to the results.
Furthermore, each industry corresponds to an industry keyword rule, when the Boolean operation result of the industry keyword rule is true, the industry is judged to belong to, and when the result is false, the industry is judged not to belong to.
Furthermore, each industry corresponds to a plurality of industry keyword rules, when the Boolean operation result of one industry keyword rule is true, the industry is judged to belong to, and when all the results are false, the industry is judged not to belong to.
Further, the industry keyword rule is a boolean operation among a plurality of keywords.
Further, the emotion score of the text data under each industry is calculated by the following steps:
dividing the text data into a plurality of sentences, identifying emotion words and main words contained in each sentence according to industry emotion dictionaries corresponding to industries to which the text data belong, and calculating emotion Score of each sentencesentence：
Wherein the subscript w denotes the affective word, ScorewExpressing the emotion score of the emotion word w recorded in the industry emotion dictionary, dwRepresenting the distance between the main words and the emotional words w;
score according to emotion Score of each sentencesentenceAnd the text length of the text data, calculating the emotion Score of the text datacontent：
Wherein: len (a)contentIndicating the text length of the text data.
Further, the method for obtaining the positive and negative emotion analysis result of the text data according to the emotion score of the text data comprises the following steps:
when the text data only belongs to one industry, if the emotion score is positive, the text data is positive emotion; when the emotion score is negative, negative emotion is detected; when the emotion score is zero, the emotion score is a middle emotion;
and when the text data belongs to at least two industries, calculating the emotion score of each industry, and judging whether the text data belongs to positive emotion, intermediate emotion or negative emotion according to a voting method.
Further, the machine learning model is a BilSTM deep cycle network model.
The positive and negative emotion analysis terminal device comprises a processor, a memory and a computer program which is stored in the memory and can run on the processor, wherein the processor executes the computer program to realize the steps of the method of the embodiment of the invention.
A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to an embodiment of the invention as described above.
According to the invention, by adopting the technical scheme, the network text is divided and treated by an emotion analysis method based on the fusion of the industry emotion words and the machine learning model, and the industry field data (such as financial and economic) is analyzed by an algorithm based on the industry emotion dictionary, only one set of industry keyword rules and the industry emotion dictionary need to be maintained, different industries load different emotion dictionaries and introduce main word and word spacing, so that the emotion analysis effect of the industry text is effectively improved; the general field text adopts a machine learning model algorithm of the current process, and the overall emotion analysis effect is improved.
Drawings
Fig. 1 is a flowchart illustrating a first embodiment of the present invention.
Fig. 2 is a flowchart illustrating the operation of the BiLSTM deep-loop network model according to an embodiment of the present invention.
Detailed Description
To further illustrate the various embodiments, the invention provides the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the embodiments. Those skilled in the art will appreciate still other possible embodiments and advantages of the present invention with reference to these figures.
The invention will now be further described with reference to the accompanying drawings and detailed description.
The first embodiment is as follows:
the embodiment of the invention provides a positive and negative emotion analysis method, as shown in FIG. 1, the method comprises the following steps:
s1: and establishing and maintaining industry keyword rules and industry emotion dictionaries corresponding to different industries.
The industry emotion dictionary comprises main words capable of representing the type of the industry, for example, the main words of the police-related industry comprise: police, jc, traffic police, etc., and emotional words that can represent the industry's emotions.
Each industry may correspond to one or more industry keyword rules, each industry keyword rule is a boolean operation among a plurality of keywords, and the format of the industry keyword rule set in this embodiment is: word 1| | word 2| | | word 3& & word 4& & & |. The word 5, wherein: | l represents a logical OR operation, & & represents a logical AND operation, |! Indicating a logical not operation, the logical operators may be assigned operation priorities with "()".
S2: judging whether the text data to be analyzed contains keywords contained in the industry keyword rule, and if yes, entering S3; otherwise, the process proceeds to S4.
In this embodiment, whether the text data contains a keyword is determined by an AC automaton algorithm.
S3: and after judging all industries to which the text data belongs according to the industries to which the keywords belong and the industry keyword rules corresponding to the industries, calculating the emotion score of each industry according to the industry emotion dictionary corresponding to each industry, and further obtaining the emotion positive and negative analysis result of the text data.
(1) Industry judgment
The method for judging all industries to which the text data belongs comprises the following steps: and calculating Boolean operation results of the industry keyword rules corresponding to each industry according to all keywords contained in the text data, and judging whether the industry belongs to the industry or not according to the results.
In this embodiment, the keywords in the industry keyword rule are replaced with "true" or "false", the keywords included in the text data are set as true, and the keywords not included are set as false, if the text data includes word 1, word 3, and word 5, and does not include word 2 and word 4, the rule is rewritten as: true | | false | | true & & false & &! And performing Boolean operation on the rewritten result to obtain a final result. In this embodiment, a high-efficiency FEL boolean operation engine is used to perform boolean operations.
And when each industry corresponds to an industry keyword rule, judging that the industry belongs to the industry when the Boolean operation result of the industry keyword rule is true, and judging that the industry does not belong to the industry when the result is false.
And when each industry corresponds to at least two industry keyword rules, judging that the industry belongs to when the Boolean operation result of one industry keyword rule is true, and judging that the industry does not belong to when all the results are false.
(2) Sentiment score calculation
The emotion score of the text data under each industry is calculated by the following steps:
A. dividing the text data into a plurality of sentences, identifying emotion words and main words contained in each sentence according to industry emotion dictionaries corresponding to industries to which the text data belong, and calculating emotion Score of each sentencesentence：
Wherein the subscript w denotes the affective word, ScorewExpressing the emotion score of the emotion word w recorded in the industry emotion dictionary, dwAnd representing the distance between the main words and the emotional words w.
The closer the sentiment word is to the subject word in the sentence, the higher the weight. If the long sentence does not contain the main word, the weight of the emotion word is 0 or a value smaller than the weight containing the main word, and is preferably set to 0 in this embodiment.
In this embodiment, the sentence is divided into passes. "to perform the division.
B.Score according to emotion Score of each sentencesentenceAnd the text length of the text data, calculating the emotion Score of the text datacontent：
Wherein: len (a)contentIndicating the text length of the text data.
(3) Determination of positive and negative sensitivity analysis results
When the emotion score is positive, the emotion score is usually positive emotion, and the probability of representing the emotion score as positive is higher;
when the emotion score is negative, negative emotion is common, and the smaller the emotion score is, the higher the probability of representing a positive emotion is.
Therefore, in this embodiment, a positive emotion threshold and a negative emotion threshold are set, and for each industry, when the emotion score of the text data is greater than the positive emotion threshold, the text data is a positive emotion, and when the emotion score is less than the negative emotion threshold, the text data is a negative emotion, and the text data is an intermediate emotion between the positive emotion threshold and the negative emotion threshold.
When the text data belong to at least two industries, judging through a voting method, namely when the number of industries with positive emotions of the text data belonging to the industry is larger than that of industries with negative emotions of the text data belonging to the industry, judging as the positive emotions; when the number of industries with positive emotions of the text data belonging to the industry is smaller than that of industries with negative emotions of the text data belonging to the industry, the text data is negative emotion; and when the number of industries with positive emotions of the text data belonging to the industry is equal to the number of industries with negative emotions of the text data belonging to the industry, the text data is the middle emotion.
S4: and obtaining the positive and negative emotion analysis result of the text data through the trained machine learning model.
The training is performed through a large amount of text data with labels searched on the Internet.
The machine learning model in this embodiment employs a BilSTM deep cycle network model.
The BilSTM deep circulation network model combines the word order and the deep semantic information of the text in the positive direction and the negative direction, and solves the long-term dependence problem through an LSTM gating result, and is a structure which is most widely used in the field of natural language processing at present.
As shown in fig. 2, in this embodiment, word segmentation is performed on the text data, then each word is represented as a dense vector through an embedding layer, then two hidden layer vectors are obtained through positive and negative LSTM network modules, the two hidden layer vectors are concat for splicing, and finally, an emotion positive and negative analysis result of the text data is obtained through a softmax layer.
According to the emotion analysis method based on the fusion of the industry emotion words and the machine learning model, the network texts are divided and treated, the industry field data (such as financial and economic) is analyzed through an algorithm based on an industry emotion dictionary, only one set of industry keyword rules and the industry emotion dictionary need to be maintained, different industries load different emotion dictionaries and introduce main word and word intervals, and the emotion analysis effect of the industry texts is effectively improved; the general field text adopts a machine learning model algorithm of the current process, and improves the overall emotion analysis effect based on algorithm analysis of a BilSTM depth model. The method has the following beneficial effects:
(1) the method has the advantages that the industry keyword rule is introduced, the text is divided into specific industry categories, the emotion words of specific industries are loaded, the purpose of correlation analysis of the text and the industry emotion words is achieved, the emotion analysis accuracy of specific industries is improved, the emotion words and industry keyword rule of related industries can be added conveniently in the later stage according to the actual effect, and the classification effect is improved.
(2) A BilSTM deep cycle network model is introduced, the model combines the word order information of the text in the positive direction and the negative direction, the long-term dependence problem is effectively solved, and the prediction effect of the model is improved. The method can find related corpus training models on the Internet and initial models in related general fields, and can optimize the training models by using a transfer learning technology on the basis of the original models according to continuous sample accumulation in the later period, so that the model prediction effect is improved.
(3) The method based on the industry emotion words and the base machine learning model is fused, the industry keyword rules are hit and predicted through the emotion dictionary, and other data are predicted through the machine learning model.
Example two:
the invention also provides positive and negative emotion analysis terminal equipment, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the steps in the method embodiment of the first embodiment of the invention are realized when the processor executes the computer program.
Further, as an executable scheme, the positive and negative emotion analysis terminal device may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The positive and negative emotion analysis terminal equipment can comprise, but is not limited to, a processor and a memory. Those skilled in the art will understand that the above-mentioned positive and negative emotion analysis terminal device is only an example of a positive and negative emotion analysis terminal device, and does not constitute a limitation on the positive and negative emotion analysis terminal device, and may include more or less components than the above, or combine some components, or different components, for example, the positive and negative emotion analysis terminal device may further include an input/output device, a network access device, a bus, etc., which is not limited in this embodiment of the present invention.
Further, as an executable solution, the processor may be a Central Processing Unit (CPU), other general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, and the like. The general processor can be a microprocessor or the processor can be any conventional processor and the like, the processor is a control center of the positive and negative emotion analysis terminal equipment, and various interfaces and lines are used for connecting all parts of the positive and negative emotion analysis terminal equipment.
The memory can be used for storing the computer program and/or the module, and the processor can realize various functions of the positive and negative emotion analysis terminal device by running or executing the computer program and/or the module stored in the memory and calling the data stored in the memory. The memory can mainly comprise a program storage area and a data storage area, wherein the program storage area can store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of the mobile phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
The invention also provides a computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the above-mentioned method of an embodiment of the invention.
The positive and negative emotion analysis terminal device integrated module/unit can be stored in a computer readable storage medium if it is implemented in the form of a software functional unit and sold or used as a separate product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-only Memory (ROM ), Random Access Memory (RAM), software distribution medium, and the like.
While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A positive and negative emotion analysis method is characterized by comprising the following steps:
s1: establishing and maintaining an industry keyword rule and an industry emotion dictionary corresponding to different industries, wherein the industry emotion dictionary comprises main words capable of expressing the type of the industry and emotion words capable of expressing the industry emotion;
s2: judging whether the text data to be analyzed contains keywords contained in the industry keyword rule, and if yes, entering S3; otherwise, go to S4;
s3: judging all industries to which the text data belongs according to the industries to which the keywords belong and the industry keyword rules corresponding to the industries, and then calculating the emotion score of each industry according to the industry emotion dictionary corresponding to each industry so as to obtain the emotion positive and negative analysis result of the text data;
s4: and obtaining the positive and negative emotion analysis result of the text data through the trained machine learning model.
2. The positive-negative emotion analysis method according to claim 1, characterized in that: the method for judging all industries to which the text data belongs comprises the following steps: and calculating Boolean operation results of the industry keyword rules corresponding to each industry according to all keywords contained in the text data, and judging whether the industry belongs to the industry or not according to the results.
3. The positive-negative emotion analysis method according to claim 2, characterized in that: each industry corresponds to an industry keyword rule, when the Boolean operation result of the industry keyword rule is true, the industry is judged to belong to, and when the result is false, the industry is judged not to belong to.
4. The positive-negative emotion analysis method according to claim 2, characterized in that: each industry corresponds to at least two industry keyword rules, when the Boolean operation result of one industry keyword rule is true, the industry is judged to belong to, and when all the results are false, the industry is judged not to belong to.
5. The positive-negative emotion analysis method according to claim 1, characterized in that: the industry keyword rule is a boolean operation among a plurality of keywords.
6. The positive-negative emotion analysis method according to claim 1, characterized in that: the emotion score of the text data under each industry is calculated by the following steps:
dividing the text data into a plurality of sentences, identifying emotion words and main words contained in each sentence according to industry emotion dictionaries corresponding to industries to which the text data belong, and calculating emotion Score of each sentencesentence：
Wherein the subscript w denotes the affective word, ScorewExpressing the emotion score of the emotion word w recorded in the industry emotion dictionary, dwRepresenting the distance between the main words and the emotional words w;
score according to emotion Score of each sentencesentenceAnd the text length of the text data, calculating the emotion Score of the text datacontent：
Wherein: len (a)contentIndicating the text length of the text data.
7. The positive-negative emotion analysis method according to claim 1, characterized in that: the method for obtaining the positive and negative emotion analysis result of the text data according to the emotion score of the text data comprises the following steps:
when the text data only belongs to one industry, judging that the text data belongs to positive emotion, middle emotion or negative emotion according to the size relation between the emotion score and a set threshold value;
and when the text data belongs to at least two industries, calculating the emotion score of each industry, and judging whether the text data belongs to positive emotion, intermediate emotion or negative emotion according to a voting method.
8. The positive-negative emotion analysis method according to claim 1, characterized in that: the machine learning model is a BilSTM deep cycle network model.
9. The utility model provides a positive negative face emotion analysis terminal equipment which characterized in that: comprising a processor, a memory and a computer program stored in the memory and running on the processor, the processor implementing the steps of the method according to any one of claims 1 to 8 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911171315.1A CN110888983B (en) | 2019-11-26 | 2019-11-26 | Positive and negative emotion analysis method, terminal equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911171315.1A CN110888983B (en) | 2019-11-26 | 2019-11-26 | Positive and negative emotion analysis method, terminal equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110888983A true CN110888983A (en) | 2020-03-17 |
| CN110888983B CN110888983B (en) | 2022-07-15 |
Family
ID=69748748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911171315.1A Active CN110888983B (en) | 2019-11-26 | 2019-11-26 | Positive and negative emotion analysis method, terminal equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110888983B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111815426A (en) * | 2020-09-11 | 2020-10-23 | 深圳司南数据服务有限公司 | Data processing method and terminal related to financial investment and research |
| CN113361267A (en) * | 2021-06-29 | 2021-09-07 | 招商局金融科技有限公司 | Sample data generation method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150302304A1 (en) * | 2014-04-17 | 2015-10-22 | XOcur, Inc. | Cloud computing scoring systems and methods |
| CN107301171A (en) * | 2017-08-18 | 2017-10-27 | 武汉红茶数据技术有限公司 | A kind of text emotion analysis method and system learnt based on sentiment dictionary |
| CN108170685A (en) * | 2018-01-29 | 2018-06-15 | 浙江省公众信息产业有限公司 | Text emotion analysis method, device and computer readable storage medium |
| CN109165298A (en) * | 2018-08-15 | 2019-01-08 | 上海文军信息技术有限公司 | A kind of text emotion analysis system of autonomous upgrading and anti-noise |
| CN109858026A (en) * | 2019-01-17 | 2019-06-07 | 深圳壹账通智能科技有限公司 | Text emotion analysis method, device, computer equipment and storage medium |
-
2019
- 2019-11-26 CN CN201911171315.1A patent/CN110888983B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150302304A1 (en) * | 2014-04-17 | 2015-10-22 | XOcur, Inc. | Cloud computing scoring systems and methods |
| CN107301171A (en) * | 2017-08-18 | 2017-10-27 | 武汉红茶数据技术有限公司 | A kind of text emotion analysis method and system learnt based on sentiment dictionary |
| CN108170685A (en) * | 2018-01-29 | 2018-06-15 | 浙江省公众信息产业有限公司 | Text emotion analysis method, device and computer readable storage medium |
| CN109165298A (en) * | 2018-08-15 | 2019-01-08 | 上海文军信息技术有限公司 | A kind of text emotion analysis system of autonomous upgrading and anti-noise |
| CN109858026A (en) * | 2019-01-17 | 2019-06-07 | 深圳壹账通智能科技有限公司 | Text emotion analysis method, device, computer equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 丁蔚: "基于词典和机器学习组合的情感分析", 《中国优秀硕士论文全文数据库 信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111815426A (en) * | 2020-09-11 | 2020-10-23 | 深圳司南数据服务有限公司 | Data processing method and terminal related to financial investment and research |
| CN111815426B (en) * | 2020-09-11 | 2020-12-15 | 深圳司南数据服务有限公司 | Data processing method and terminal related to financial investment and research |
| CN113361267A (en) * | 2021-06-29 | 2021-09-07 | 招商局金融科技有限公司 | Sample data generation method, device, equipment and storage medium |
| CN113361267B (en) * | 2021-06-29 | 2024-02-09 | 招商局金融科技有限公司 | Sample data generation method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110888983B (en) | 2022-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109858010B (en) | Method and device for recognizing new words in field, computer equipment and storage medium | |
| CN108446271B (en) | Text emotion analysis method of convolutional neural network based on Chinese character component characteristics | |
| CN110083832B (en) | Article reprint relation identification method, device, equipment and readable storage medium | |
| CN112101041A (en) | Entity relationship extraction method, device, equipment and medium based on semantic similarity | |
| CN111339260A (en) | BERT and QA thought-based fine-grained emotion analysis method | |
| US10915756B2 (en) | Method and apparatus for determining (raw) video materials for news | |
| CN111832290A (en) | Model training method and device for determining text relevancy, electronic equipment and readable storage medium | |
| CN112926308B (en) | Method, device, equipment, storage medium and program product for matching text | |
| CN110888983B (en) | Positive and negative emotion analysis method, terminal equipment and storage medium | |
| Najafi et al. | Text-to-Text Transformer in Authorship Verification Via Stylistic and Semantical Analysis. | |
| CN113806493A (en) | Entity relationship joint extraction method and device for Internet text data | |
| Shi et al. | A Word2vec model for sentiment analysis of weibo | |
| CN114722198A (en) | Method, system and related device for determining product classification code | |
| Jayakody et al. | Sentiment analysis on product reviews on twitter using Machine Learning Approaches | |
| CN116029280A (en) | Method, device, computing equipment and storage medium for extracting key information of document | |
| CN113377910A (en) | Emotion evaluation method and device, electronic equipment and storage medium | |
| CN108021609B (en) | Text emotion classification method and device, computer equipment and storage medium | |
| Kavitha et al. | A review on machine learning techniques for text classification | |
| Arbaatun et al. | Hate speech detection on Twitter through Natural Language Processing using LSTM model | |
| CN115640399A (en) | Text classification method, device, equipment and storage medium | |
| Suhasini et al. | A Hybrid TF-IDF and N-Grams Based Feature Extraction Approach for Accurate Detection of Fake News on Twitter Data | |
| CN111191455A (en) | Legal provision prediction method in traffic accident damage compensation | |
| CN115169345A (en) | Training method, device and equipment for text emotion analysis model and storage medium | |
| Braoudaki et al. | Hybrid data driven and rule based sentiment analysis on Greek text | |
| CN114969341A (en) | Fine-grained emotion analysis method and device for catering industry comments |
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 |