CN110929528B - Method, device, server and storage medium for analyzing emotion of sentence - Google Patents
Method, device, server and storage medium for analyzing emotion of sentence Download PDFInfo
- Publication number
- CN110929528B CN110929528B CN201911147402.3A CN201911147402A CN110929528B CN 110929528 B CN110929528 B CN 110929528B CN 201911147402 A CN201911147402 A CN 201911147402A CN 110929528 B CN110929528 B CN 110929528B
- Authority
- CN
- China
- Prior art keywords
- words
- word
- vocabulary
- vector
- information vector
- 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
Classifications
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a method, a device, a server and a storage medium for analyzing emotion of a sentence, and belongs to the technical field of natural language processing. The method comprises the following steps: determining first word vectors corresponding to a plurality of words included in a sentence to be analyzed; processing the first word vectors corresponding to the plurality of words, the dependency relationships among the plurality of words and the types of the dependency relationships based on the graph rolling network to obtain second word vectors corresponding to the plurality of words; and determining the emotion type of at least one first target vocabulary in the plurality of vocabularies based on the second word vectors corresponding to the plurality of vocabularies. Because the second word vector corresponding to the plurality of words is determined based on the graph-convolution network, the dependency relationship among the plurality of words and the type of each dependency relationship are considered, even if the sentence to be analyzed is long, the relationship among the words can be enhanced through the dependency relationship among the words, and therefore the accuracy of the analysis result of the emotion type of each entity word is improved.
Description
Technical Field
The present application relates to the field of natural language processing technologies, and in particular, to a method, an apparatus, a server, and a storage medium for emotion analysis of a sentence.
Background
Emotion analysis is a common scenario in natural language processing and can be used for guiding iterative updating of products or improving service level and the like. For example, for meal evaluation in daily life, emotion types corresponding to the entities are identified and analyzed through entity identification of evaluation contents, emotion information of a user on multiple dimensions such as dish taste, dish feeding time, service attitude and meal environment can be obtained, and the quality degree of each aspect (aspect-level) at the present stage can be determined based on the emotion information, so that how to improve is clear.
In the related art, there is an emotion analysis method based on deep learning. The specific implementation steps are as follows: firstly, carrying out word vector coding on each word included in a sentence to be analyzed, then carrying out feature extraction based on a convolutional neural network, and finally outputting the probability that each word belongs to an entity word, thereby obtaining the probability of the emotion category to which each entity word belongs.
The problem in the related art is that when the sentence to be analyzed is long, the distance between the vocabularies with the association relationship in the sentence is large, thereby affecting the accuracy of the analysis result.
Disclosure of Invention
The embodiment of the application provides a method, a device, a server and a storage medium for emotion analysis of sentences, which are used for solving the problem that when sentences to be analyzed are longer, the distance between vocabularies with association relations in the sentences is larger, so that the accuracy of analysis results is affected. The technical scheme is as follows:
in one aspect, a method for emotion analysis of a sentence is provided, including:
determining first word vectors corresponding to a plurality of words included in a sentence to be analyzed;
processing the first word vectors corresponding to the plurality of words, the dependency relations among the plurality of words and the types of the dependency relations based on a graph rolling network to obtain second word vectors corresponding to the plurality of words;
and determining the emotion type of at least one first target vocabulary in the plurality of vocabularies based on the second word vectors corresponding to the plurality of vocabularies.
In another aspect, an apparatus for emotion analysis of a sentence is provided, including:
the determining module is used for determining first word vectors corresponding to a plurality of words included in the sentences to be analyzed;
the processing module is used for processing the first word vectors corresponding to the plurality of words, the dependency relations among the plurality of words and the types of the dependency relations based on the graph rolling network to obtain second word vectors corresponding to the plurality of words;
The determining module is further configured to determine, based on the second word vectors corresponding to the plurality of words, an emotion type to which at least one first target word in the plurality of words belongs.
In one possible implementation manner, the determining module is further configured to segment the sentence to be analyzed to obtain a plurality of vocabularies; and for any vocabulary, splicing word vectors corresponding to the first meaning of the vocabulary with word vectors corresponding to the second meaning of the vocabulary to obtain the first word vectors corresponding to the vocabulary.
In another possible implementation manner, the processing module is further configured to obtain a dependency relationship between the plurality of vocabularies; determining a relation matrix according to the dependency relation, wherein elements in the relation matrix are used for representing whether the dependency relation exists among the vocabularies; for any dependency, acquiring a relationship vector corresponding to the dependency according to the type of the dependency; and processing the relation matrix and the plurality of relation vectors based on the graph rolling network to obtain second word vectors corresponding to the plurality of words.
In another possible implementation manner, the processing module is further configured to obtain, for any vocabulary of the plurality of vocabularies, a first word vector corresponding to at least one vocabulary having a dependency relationship with the vocabulary; and inputting the relation matrix, the plurality of relation vectors and the first word vector corresponding to the at least one vocabulary into the graph rolling network to obtain the second word vector corresponding to the vocabulary.
In another possible implementation manner, the determining module is further configured to process first word vectors corresponding to the plurality of words based on a first convolutional neural network, so as to obtain third word vectors corresponding to the plurality of words, where the third word vectors are in one-to-one correspondence with the second word vectors; splicing the corresponding second word vector and the third word vector to obtain a fourth word vector corresponding to the plurality of words; and determining the emotion type of at least one first target vocabulary in the plurality of vocabularies based on the fourth word vector corresponding to the plurality of vocabularies.
In another possible implementation manner, the determining module is further configured to determine a first information vector corresponding to the plurality of words and a second information vector corresponding to the plurality of words according to a second word vector corresponding to the plurality of words, where the first information vector is used to indicate whether any word in the plurality of words is a first target word, and the second information vector is used to indicate an emotion type to which each word in the plurality of words belongs; for each vocabulary, when the probability that the vocabulary belongs to the first target vocabulary is determined to be larger than the target probability threshold according to the first information vector corresponding to the vocabulary, the probability that the vocabulary belongs to each emotion type is determined according to the second information vector corresponding to the vocabulary.
In another possible implementation manner, the determining module is further configured to recode second word vectors corresponding to the plurality of words based on a second convolutional neural network, so as to obtain first information vectors corresponding to the plurality of words; recoding second word vectors corresponding to the plurality of words based on a third convolutional neural network to obtain second information vectors corresponding to the plurality of words; splicing the first information vector and the second information vector to obtain a third information vector; and respectively recoding the third information vector based on the second convolutional neural network and the third convolutional neural network to obtain an updated first information vector and a second information vector.
In another possible implementation, the apparatus further includes:
the splicing module is used for splicing the updated first information vector and the updated second information vector to obtain an updated third information vector;
the coding module is used for recoding the updated third information vector again based on the second convolutional neural network and the third convolutional neural network respectively;
and the updating module is used for repeating the splicing and recoding processes until the updating stopping condition is reached.
In another possible implementation manner, the determining module is further configured to obtain interaction transfer information obtained by the second convolutional neural network according to a second word vector corresponding to the plurality of words, where the interaction transfer information is used to indicate a word belonging to a second target word in the plurality of words; recoding the third information vector based on the third convolutional neural network and the interactive transfer information to obtain an updated second information vector; and recoding the third information vector based on the second convolutional neural network to obtain an updated first information vector.
In another aspect, a server is provided that includes a processor and memory for storing at least one piece of program code that is loaded and executed by the processor to perform operations performed in a method for emotion analysis of a sentence in an embodiment of the present application.
In another aspect, a storage medium having stored therein at least one piece of program code for execution by a processor and to perform a method for emotion analysis of a sentence in an embodiment of the present application is provided.
The technical scheme provided by the embodiment of the application has the beneficial effects that:
in the embodiment of the application, because the dependency relationship among the words and the type of each dependency relationship are also considered when the second word vector corresponding to the plurality of words is determined for the first word vector corresponding to the plurality of words based on the graph rolling network, even if the sentence to be analyzed is longer, the relationship among the words can be enhanced through the dependency relationship among the words, thereby improving the accuracy of the analysis result of the emotion type of each entity word.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of a sentence emotion analysis system according to an embodiment of the present application;
FIG. 2 is a flow chart of a method for emotion analysis for sentences provided by an embodiment of the present application;
FIG. 3 is a diagram illustrating one vocabulary dependency relationship according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a method for emotion analysis of sentences according to an embodiment of the present application;
FIG. 5 is a block diagram of an apparatus for emotion analysis for sentences provided by an embodiment of the present application;
fig. 6 is a schematic structural diagram of a server according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the embodiments of the present application will be described in further detail with reference to the accompanying drawings.
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the accompanying claims.
The embodiment of the application relates to a scene requiring emotion analysis of sentences. Accordingly, the method can be applied to the scene in daily life, for example, a customer can post meal evaluation after taking a meal, a customer can post meal delivery evaluation after receiving takeaway, and a customer can post commodity evaluation after purchasing commodity. The evaluation has positive evaluation, such as pleasant meal, good environment, quick meal delivery, good commodity use and the like, has negative evaluation, such as difficult eating of dishes, high price, slow dish feeding, cool take-out, rough commodity work and the like, and has neutral evaluation, such as running, and the like. Through emotion analysis of sentences in the evaluation, dishes and services can be improved by a restaurant, or food delivery service can be improved by a takeaway service provider, or commodity upgrading and updating can be facilitated. Correspondingly, the method can also be applied to scenes in the Internet, for example, a user can comment on a movie or a television play, and the favorite degree of the user on the movie or the television play can be known through analysis of comment content, so that the playing time is designated; users can also publish views on the popular events, and related departments can be helped to know the views of citizens on the popular events through analysis of the view contents; the user can share the mood of the user, and the mood of the user can be known through analysis of mood content. Accordingly, the method can also be applied to other scenes, and is not listed here. Therefore, the embodiment of the application provides a method for analyzing emotion of a sentence.
The following briefly describes the main flow of the method for emotion analysis of sentences provided in the embodiment of the present application. First, first word vectors corresponding to a plurality of words included in a sentence to be analyzed are determined, and then the first word vectors, the dependency relationships among the plurality of words and the types of the dependency relationships are processed based on a graph convolution network, so that second word vectors corresponding to the plurality of words are obtained. And finally, determining the emotion type of at least one first target vocabulary in the plurality of vocabularies based on the second word vectors corresponding to the plurality of vocabularies, and realizing emotion analysis of all aspects of the sentences to be analyzed.
Fig. 1 is a block diagram of a sentence emotion analysis system 100 according to an embodiment of the present application, where the sentence emotion analysis system 100 may be used to implement emotion analysis on a sentence, and includes: a plurality of terminals 110 and a sentence emotion analysis platform 120.
Terminal 110 may be connected to sentence emotion analysis platform 120 via a wireless network or a wired network. Terminal 110 may be at least one of a smart phone, a video camera, a desktop computer, a tablet computer, an MP4 player, and a laptop portable computer. The terminal 110 is installed and operated with an application program including a sentence uploading function, such as a take-away program, a shopping program, a comment program, and the like. Illustratively, the terminal 110 may be a terminal used by a user, and an application program run by the terminal has an account number of the user logged in.
Statement emotion analysis platform 120 includes at least one of a server, a plurality of servers, and a cloud computing platform. Statement emotion analysis platform 120 is used to provide background services for the application program described above, including statement upload functions, such as statement segmentation, vocabulary extraction, and statement emotion analysis. In one possible manner, the sentence emotion analysis platform 120 includes: the system comprises an access server, an emotion analysis server and a database. The access server is used for providing access services for the terminal 110. The emotion analysis server is used for sentence segmentation, vocabulary extraction, sentence emotion analysis and the like. The emotion analysis server may be one or more, and when the emotion analysis servers are multiple, there are at least two emotion analysis servers for providing different services, and/or there are at least two emotion analysis servers for providing the same service, such as providing the same service in a load balancing manner or providing the same service in a manner of a master server and a mirror server, which embodiments of the present application are not limited. The database is used for storing word vectors, emotion types and the like corresponding to the vocabulary meanings. The statement to be analyzed is data information which the user is authorized to collect.
Terminal 110 may refer broadly to one of a plurality of terminals, with the present embodiment being illustrated only by terminal 110. Those skilled in the art will recognize that the number of terminals may be greater or lesser. For example, the number of the terminals may be only one, or the number of the terminals may be tens or hundreds, or more, and the program service system may further include other terminals. The embodiment of the application does not limit the number and the types of the terminals.
Fig. 2 is a flowchart of a method for emotion analysis of a sentence according to an embodiment of the present application, as shown in fig. 2. The method comprises the following steps:
201. the server determines first word vectors corresponding to a plurality of words included in the sentence to be analyzed.
In the embodiment of the application, the server can acquire at least one statement to be analyzed from at least one application program for uploading the statement. For any sentence to be analyzed, if the sentence needs to be segmented, such as a Chinese sentence, the server can segment the sentence to obtain a plurality of words; if the sentence does not need word segmentation, such as an English sentence, the server can directly acquire a plurality of words included in the sentence. The server may store a correspondence between each word and a word vector, and determine a first word vector corresponding to the plurality of words according to the correspondence.
In one possible implementation manner, since the vocabulary may have different meanings in different domains, the server may determine the first word vector corresponding to the plurality of vocabularies according to the basic meaning of the vocabulary and the special meaning of the vocabulary in the target domain. Accordingly, the steps may be: the server may segment the sentence to be analyzed to obtain a plurality of words. For any vocabulary, the server may obtain a word vector corresponding to the first meaning of the vocabulary and a word vector corresponding to the second meaning of the vocabulary, and splice the word vector corresponding to the first meaning of the vocabulary and the word vector corresponding to the second meaning of the vocabulary to obtain a first word vector corresponding to the vocabulary, thereby obtaining first word vectors corresponding to a plurality of vocabularies. The server splices word vectors corresponding to different meanings of the vocabulary, so that the first word vector corresponding to the obtained vocabulary is more relevant to the meaning of the vocabulary.
Where the first meaning of a word refers to the basic meaning of the word, for example the basic meaning of the word "lean" is that the component is lean, corresponding to "rich". The second meaning of the vocabulary means a special meaning of the vocabulary in a target field, and the target field may be a field to which a scene to which the sentence is applied belongs, such as a travel field, a shopping field, a catering field, an art field, and the like. Taking the catering field as an example, the special meaning of the light food in the catering field is light taste, and the light food corresponds to the finished food. The server may store word vectors corresponding to different meanings of the words, each word corresponding to at least one word vector.
It should be noted that, the application program including the sentence uploading function may further include other functions according to the application scenario. For example, when the application is an application applied to a shopping scenario, the application may further include functions such as searching for merchandise, adding shopping carts, order payment, and order evaluation. For another example, when the application program is an application program applied to a dining scene, the application program can further include functions of menu display, displaying a meal delivery position, meal evaluation and the like.
202. The server processes the first word vectors corresponding to the words, the dependency relations among the words and the types of the dependency relations based on the graph rolling network to obtain second word vectors corresponding to the words.
In the embodiment of the application, the server can obtain the second word vectors corresponding to the plurality of words based on the graph rolling network. The graph rolling network can be a DREGCN (Dependency Relation Embedded Graph Convolutional Networks, dependency relationship embedded graph rolling network), and compared with the traditional GCN (Graph Convolutional Networks, graph rolling network), the relationship vector corresponding to the dependency relationship among the vocabularies is added in the DREGCN.
Before introducing the DREGCN of the embodiments of the present application, we will first introduce a conventional GCN. The conventional GCN is one of GNNs (Graph Neural Networks, graphic neural networks), and a word vector corresponding to any word at time t can be calculated by the following formula (1).
Where n represents the vocabulary number, reLU represents the activation function, A ij Representing a relationship matrix, W g The parameters of the linear transformation are represented as,word vector representing the jth vocabulary of layer i, b g Representing the bias parameters.
The DREGCN in the embodiment of the application is improved on the traditional GCN, and the word vector corresponding to any word at the time t can be calculated by the following formula (2).
Wherein N represents the number of words, N represents the number of types of dependency relationships between words, reLU represents the activation function, A ij Representing a relationship matrix, wr representing a linear transformation parameter,representing a word vector corresponding to the jth vocabulary of the first layer, b r Representing the bias parameter, R < k ]]A relationship vector corresponding to the type representing the kth dependency relationship, "; "means vector concatenation.
The above W r And b r The method can be obtained through model training, and the mode of model training is not limited.
In one possible implementation manner, the server may determine a relationship matrix according to the dependency relationship among the plurality of words, determine a relationship vector corresponding to the dependency relationship according to the type to which the dependency relationship belongs, and then input the first word vector corresponding to the plurality of words, the relationship matrix and the relationship vector corresponding to each dependency relationship into the graph convolution network to obtain the second word vector corresponding to the plurality of words. Accordingly, this step may be implemented by the following sub-steps 2021 to 2024:
2021. The server can acquire the dependency relationship among a plurality of vocabularies;
a sentence may include a plurality of words, each having a respective part of speech, such as nouns, pronouns, adjectives, prepositions, etc., and each having a respective effect, such as subject, predicate, object, and subject, etc. The dependency relationship among the words in the sentence can form a multi-layer tree structure, and the words are connected according to the dependency relationship.
For example, take the phrase "coffee is a cheaper item than a high cost ham sandwich" as an example. The words "coffee", "yes", "one", "comparison", "high price", "ham", "sandwich", "cheap" and "things" can be obtained after the word segmentation. The dependency between the words can be seen in fig. 3. FIG. 3 is a diagram illustrating a vocabulary dependency relationship according to an embodiment of the present application. Words with dependency relationships are connected through line segments, such as "yes" and "coffee", "things" with dependency relationships, "things" and "is", "one", "cheap", "ratio" with dependency relationships, "ratio" and "sandwich", "things" with dependency relationships, "sandwich" and "ratio", "high price", "ham" with dependency relationships.
2022. And the server determines a relation matrix according to the dependency relation, wherein elements in the relation matrix are used for representing whether the dependency relation exists among the vocabularies.
After the server acquires the dependency relationship, a relationship matrix can be constructed according to the dependency relationship, wherein the number of rows and columns of the relationship matrix are equal to the number of words. When any two vocabularies have a dependency relationship, the server can set an element which represents whether the dependency relationship exists between the two vocabularies as 1 in a relationship matrix; when any two words do not have a dependency relationship, the server may set an element indicating whether there is a dependency relationship between the two words to 0 in the relationship matrix.
For example, the above description will be continued taking the phrase "coffee is a cheaper item than a high-priced ham sandwich" as an example. After obtaining the dependency relationship of each vocabulary, the server may determine the following relationship matrix a.
Wherein, the element value of the first row and the first column of the relation matrix A is 1, which indicates that 'coffee' has a dependency relation with the coffee; the element value of the second column of the first row is 1, which indicates that the 'coffee' and the 'yes' have a dependency relationship; the element value of the seventh column of the fourth row is 1, which indicates that the 'ratio' has a dependency relationship with the 'sandwich'; the element value of the ninth column of the fourth row is 1, which indicates that the 'ratio' has a dependency relationship with 'things'. And so on.
2023. For any dependency, the server may obtain a relationship vector corresponding to the dependency according to the type to which the dependency belongs.
For any of the above-identified at least one dependency, the server may identify the type to which the dependency belongs, e.g., the noun subject that "coffee" is "yes," and accordingly, the predicate that "yes" is "coffee," the compound of "ham" is "sandwich," which together constitute the adjective "ham sandwich," high-priced "is" sandwich, "and the object of the preposition that" sandwich "is" ratio. Each type of dependency corresponds to a relationship vector, and the server can determine the relationship vector corresponding to the dependency according to the type of the dependency.
In one possible implementation, the server may store a correspondence between the type of dependency and the relationship vector. The server may determine, according to the correspondence, a type to which the dependency relationship belongs, and then obtain a corresponding relationship vector.
It should be noted that, the types of the dependency relationships between the vocabularies are paired, that is, the type of the dependency relationship between the vocabulary a and the vocabulary B is a, and the type of the dependency relationship between the vocabulary B and the vocabulary a is B, for example, the noun subject that "coffee" is "yes", and correspondingly, the predicate that "coffee" is "yes".
2024. And the server processes the relation matrix and the plurality of relation vectors based on the graph rolling network to obtain second word vectors corresponding to the plurality of vocabularies.
After the server obtains the relationship matrix and the plurality of relationship vectors, for any one of the plurality of words, the server may obtain a first word vector corresponding to at least one word having a dependency relationship with the word. The server may input the relationship matrix, the plurality of relationship vectors, and the first word vector corresponding to the at least one word into a graph convolution network to obtain a second word vector corresponding to the word. Because the dependency relationship among the vocabularies is considered when the processing is performed based on the graph convolution network, the second word vector corresponding to the obtained vocabularies is more closely related to the second word vectors corresponding to other vocabularies, and therefore the accuracy of analysis results is improved.
203. The server determines a first information vector corresponding to the plurality of words and a second information vector corresponding to the plurality of words according to a second word vector corresponding to the plurality of words, wherein the first information vector is used for indicating whether any word in the plurality of words is a first target word, and the second information vector is used for indicating the emotion type of each word in the plurality of words.
In the embodiment of the present application, after obtaining the second word vectors corresponding to the plurality of words, the server may determine the first information vector and the second information vector by respectively processing the second word vectors corresponding to the plurality of words based on an aspect word extraction task and an aspect-level emotion classification task. The aspect vocabulary extraction task is used for identifying a first target vocabulary and a second target vocabulary included in the plurality of vocabularies, wherein the first target vocabulary can be entity vocabularies such as nouns and pronouns, and the second target vocabulary can be emotion vocabulary such as adjectives. The aspect-level emotion classification task may include aspect entity identification, emotion classification, and emotion word identification.
In one possible implementation manner, the aspect vocabulary extraction task may be performed by a second convolutional neural network, and the aspect-level emotion classification task may be performed by a third convolutional neural network. Accordingly, the implementation manner of this step may be: the server may recode second word vectors corresponding to the plurality of words based on the second convolutional neural network, to obtain first information vectors corresponding to the plurality of words. The server can also re-encode the second word vectors corresponding to the plurality of words based on the third convolutional neural network to obtain second information vectors corresponding to the plurality of words. The server may splice the first information vector and the second information vector to obtain a third information vector. The server may re-encode the third information vector based on the second convolutional neural network and the third convolutional neural network, respectively, to obtain an updated first information vector and a second information vector. After the server splices the obtained first information vector and the second information vector, the first information vector and the second information vector are respectively input into the second convolutional neural network and the third convolutional neural network again for recoding, so that information sharing is realized among different tasks, namely, the information of a first target vocabulary obtained by an aspect vocabulary extraction task is shared to an aspect-level emotion classification task, and the emotion type information of each vocabulary obtained by the aspect-level emotion classification task is shared to the aspect vocabulary extraction task.
In one possible implementation, the server may share information multiple times, i.e. update the first information vector and the second information vector multiple times. The corresponding steps can be as follows: the server may splice the updated first information vector and the updated second information vector to obtain an updated third information vector. The server may re-encode the updated third information vector again based on the second convolutional neural network and the third convolutional neural network, respectively. The above-described splicing and re-encoding process is repeated until a stop update condition is reached. The update stopping condition may be that the target iteration number is reached, for example, 2 times, or the first information vector and the second information vector converge, which is not limited in the embodiment of the present application.
In one possible implementation manner, the server may further send the adjective vocabulary information obtained in the aspect vocabulary extraction task to the aspect-level emotion analysis task, so that the server updates the second information vector according to the adjective vocabulary information when executing the emotion analysis task. Correspondingly, the step of obtaining the updated first information vector and the updated second information vector by the server may be: the server may obtain interaction transfer information obtained by the second convolutional neural network according to the second word vectors corresponding to the plurality of words, where the interaction transfer information is used to indicate a word belonging to the second target word in the plurality of words. The server may re-encode the third information vector based on the third convolutional neural network and the interactive transfer information to obtain an updated second information vector. The server may re-encode the third information vector based on the second convolutional neural network to obtain an updated first information vector. The server shares the information which indicates the second target vocabulary in the plurality of vocabularies, namely the information of adjective vocabulary, to an aspect-level emotion analysis task, so that a second information vector updated based on the third convolutional neural network is more accurate. Thereby improving the accuracy of the results.
When the server performs information sharing between the different tasks, the information sharing may be realized by the following formula (3). The server may splice the third information vector input last time with the obtained first information vector and the second information vector, and then re-encode the first information vector and the second information vector to obtain an updated third information vector.
Wherein, the liquid crystal display device comprises a liquid crystal display device,representing information vector corresponding to ith vocabulary after t-th information sharing, fθ re Representing recoding function->Information vector corresponding to the ith vocabulary in the aspect vocabulary extraction task is represented by +.>And the information vector corresponding to the i-th vocabulary in the aspect-level emotion analysis task is represented.
204. And for each vocabulary, when the server determines that the probability that the vocabulary belongs to the first target vocabulary is larger than the target probability threshold according to the first information vector corresponding to the vocabulary, determining the probability that the vocabulary belongs to each emotion type according to the second information vector corresponding to the vocabulary.
In the embodiment of the application, for each vocabulary, the server can perform full connection processing on the first information vector corresponding to the vocabulary to obtain the probability that the vocabulary belongs to the first target vocabulary or the second target vocabulary. When the probability that the vocabulary belongs to the first target vocabulary is larger than the target probability threshold, the server predicts emotion according to the second information vector corresponding to the vocabulary, so that the probability that the vocabulary belongs to each emotion type is obtained, and the emotion type of the vocabulary is obtained. Wherein the sum of probabilities that any word belongs to each emotion type is 1.
For example, the emotion type of the entity word "coffee" may have three emotion types of "positive", "neutral" and "negative", and in the above statement, "cheap" means that "coffee" is cheaper than "sandwich", which is an advantage of "coffee", so that the probability that the emotion type of the entity word "coffee" belongs to "positive" after emotion prediction of the entity word "coffee" is higher, and thus the emotion type of the entity word "coffee" is "positive".
It should be noted that, step 203 and step 204 are one possible implementation manner of determining, by the server, an emotion type to which at least one first target vocabulary in the plurality of vocabularies belongs based on the second word vectors corresponding to the plurality of vocabularies.
In one possible implementation manner, the server may further obtain a third word vector corresponding to the plurality of words based on the first convolutional neural network, and determine an emotion type to which at least one first target word in the plurality of words belongs based on a fourth word vector obtained by splicing the third word vector and the second word vector. The corresponding implementation steps can be as follows: the server can process the first word vectors corresponding to the plurality of words based on the first convolutional neural network to obtain third word vectors corresponding to the plurality of words, wherein the third word vectors correspond to the second word vectors one by one. The server can splice the corresponding second word vector and the corresponding third word vector to obtain a fourth word vector corresponding to the plurality of words, and the server can determine the emotion type of at least one first target word in the plurality of words based on the fourth word vector corresponding to the plurality of words. The step of determining, by the server, the emotion type to which the at least one first target vocabulary belongs based on the fourth word vector corresponding to the plurality of vocabularies may refer to the implementation manner of determining, by the server, the emotion type to which the at least one first target vocabulary belongs based on the second word vector corresponding to the plurality of vocabularies, which is not described in detail in the embodiments of the present application.
In addition, in order to more clearly show the architecture of the method for analyzing the emotion of the sentence provided by the embodiment of the present application, fig. 4 may be shown, and fig. 4 is a schematic diagram of the architecture of the method for analyzing the emotion of the sentence provided by the embodiment of the present application. Input in fig. 4 is a plurality of words obtained after word segmentation of the sentence to be analyzed. The coding layer comprises two parts, wherein the first part is to splice word vectors corresponding to first meanings of each vocabulary with word vectors corresponding to second meanings of the vocabulary to obtain the first word vectors corresponding to the vocabulary. The second part is to process the first word vectors corresponding to the plurality of words, the dependency relations among the plurality of words and the types of the dependency relations based on the graph convolution network to obtain second word vectors corresponding to the plurality of words, and of course, the second part is also to process the first word vectors corresponding to the plurality of words based on the first convolution neural network to obtain third word vectors corresponding to the plurality of words. The server can splice the corresponding second word vector and the third word vector to obtain a fourth word vector corresponding to the plurality of vocabularies. The task-oriented layer comprises a second convolutional neural networkAnd third convolutional neural network- >AE represents an aspect vocabulary extraction task and AS represents an aspect-level emotion classification task. The prediction layer is used for predicting the emotion type of the entity vocabulary.
In the embodiment of the application, because the dependency relationship among the words and the type of each dependency relationship are also considered when the second word vector corresponding to the plurality of words is determined for the first word vector corresponding to the plurality of words based on the graph rolling network, even if the sentence to be analyzed is longer, the relationship among the words can be enhanced through the dependency relationship among the words, thereby improving the accuracy of the analysis result of the emotion type of each entity word.
Compared with the related art, the method for analyzing the emotion of the sentence has the advantages that indexes such as F1 score and accuracy are greatly improved. In order to more clearly demonstrate the beneficial effects of the present application, comparative experiments were performed on the present application and related techniques. In the comparison experiment, three general data sets are respectively processed by the method and the related technology based on the related technology. The indexes of the comparison experiment are an F1 value for identifying the entity, an F1 value for identifying the emotion words, the accuracy of the aspect-level emotion classification, an F1 value for the aspect-level emotion classification and an F1 value which is extracted by the entity and has correct emotion classification. The results of the comparative experiments can be seen in table 1.
TABLE 1
As can be seen from Table 1, the method for emotion analysis of sentences provided by the application has a remarkable improvement on each index compared with the related art.
In addition, the application also compares the common GCN, DREGCN, DREGCN combined information sharing and DREGCN combined information sharing and the difference of external knowledge. And the adopted comparison index is an F1 value which is extracted by the entity and has correct emotion classification. Wherein the data set used for comparison is still the three general data sets described above. The comparison results can be seen in table 2.
TABLE 2
As can be seen from table 2, by adopting the DREGCN provided by the embodiment of the present application, compared with the common GCN, the F1 value that is extracted correctly and the emotion classification is improved significantly, and after combining the DREGCN with the information sharing and the external knowledge, the utility model is improved significantly compared with the common GCN. Wherein the external knowledge refers to a vector representation of the vocabulary in an embodiment of the application.
Fig. 5 is a block diagram of an apparatus for emotion analysis of sentences according to an embodiment of the present application. The device is configured to perform the steps when the method for interacting with a virtual object is performed, and referring to fig. 5, the device includes: a determination module 501 and a processing module 502.
The determining module is used for determining first word vectors corresponding to a plurality of words included in the sentences to be analyzed;
the processing module is used for processing the first word vectors corresponding to the plurality of words, the dependency relations among the plurality of words and the types of the dependency relations based on the graph rolling network to obtain second word vectors corresponding to the plurality of words;
the determining module is further configured to determine an emotion type to which at least one first target vocabulary belongs in the plurality of vocabularies based on the second word vectors corresponding to the plurality of vocabularies.
In one possible implementation manner, the determining module is further configured to segment the sentence to be analyzed to obtain a plurality of vocabularies; and for any vocabulary, splicing word vectors corresponding to the first meaning of the vocabulary with word vectors corresponding to the second meaning of the vocabulary to obtain first word vectors corresponding to the vocabulary.
In another possible implementation manner, the processing module is further configured to obtain a dependency relationship between a plurality of vocabularies; determining a relation matrix according to the dependency relation, wherein elements in the relation matrix are used for representing whether the dependency relation exists among the vocabularies; for any dependency, acquiring a relationship vector corresponding to the dependency according to the type of the dependency; and processing the relation matrix and the plurality of relation vectors based on the graph rolling network to obtain second word vectors corresponding to the plurality of words.
In another possible implementation manner, the processing module is further configured to obtain, for any one of the plurality of vocabularies, a first word vector corresponding to at least one vocabulary having a dependency relationship with the vocabulary; and inputting the relation matrix, the plurality of relation vectors and the first word vector corresponding to at least one vocabulary into a graph convolution network to obtain a second word vector corresponding to the vocabulary.
In another possible implementation manner, the determining module is further configured to process first word vectors corresponding to the plurality of words based on the first convolutional neural network, so as to obtain third word vectors corresponding to the plurality of words, where the third word vectors are in one-to-one correspondence with the second word vectors; splicing the corresponding second word vector and the third word vector to obtain a fourth word vector corresponding to the plurality of words; and determining the emotion type of at least one first target vocabulary in the plurality of vocabularies based on the fourth word vector corresponding to the plurality of vocabularies.
In another possible implementation manner, the determining module is further configured to determine a first information vector corresponding to the plurality of words and a second information vector corresponding to the plurality of words according to a second word vector corresponding to the plurality of words, where the first information vector is used to indicate whether any word in the plurality of words is a first target word, and the second information vector is used to indicate an emotion type to which each word in the plurality of words belongs; for each vocabulary, when the probability that the vocabulary belongs to the first target vocabulary is determined to be larger than the target probability threshold according to the first information vector corresponding to the vocabulary, the probability that the vocabulary belongs to each emotion type is determined according to the second information vector corresponding to the vocabulary.
In another possible implementation manner, the determining module is further configured to recode second word vectors corresponding to the plurality of words based on the second convolutional neural network, so as to obtain first information vectors corresponding to the plurality of words; recoding second word vectors corresponding to the plurality of words based on a third convolutional neural network to obtain second information vectors corresponding to the plurality of words; splicing the first information vector and the second information vector to obtain a third information vector; and respectively recoding the third information vector based on the second convolutional neural network and the third convolutional neural network to obtain an updated first information vector and a second information vector.
In another possible implementation, the apparatus further includes:
the splicing module is used for splicing the updated first information vector and the updated second information vector to obtain an updated third information vector;
the coding module is used for recoding the updated third information vector again based on the second convolutional neural network and the third convolutional neural network respectively;
and the updating module is used for repeating the splicing and recoding processes until the updating stopping condition is reached.
In another possible implementation manner, the determining module is further configured to obtain interaction transfer information obtained by the second convolutional neural network according to a second word vector corresponding to the plurality of words, where the interaction transfer information is used to indicate a word belonging to the second target word in the plurality of words; recoding a third information vector based on the third convolutional neural network and the interactive transfer information to obtain an updated second information vector; and recoding the third information vector based on the second convolutional neural network to obtain the updated first information vector.
In the embodiment of the application, because the dependency relationship among the words and the type of each dependency relationship are also considered when the second word vector corresponding to the plurality of words is determined for the first word vector corresponding to the plurality of words based on the graph rolling network, even if the sentence to be analyzed is longer, the relationship among the words can be enhanced through the dependency relationship among the words, thereby improving the accuracy of the analysis result of the emotion type of each word.
Fig. 6 is a schematic structural diagram of a server 600 according to an embodiment of the present application. The server 600 may be configured or configured differently, and may include one or more processors (Central Processing Units, CPU) 601 and one or more memories 602, where the memories 602 store at least one instruction that is loaded and executed by the processors 601 to implement the methods provided by the above-described method embodiments. Of course, the server may also have a wired or wireless network interface, a keyboard, an input/output interface, and other components for implementing the functions of the device, which are not described herein.
The embodiment of the application also provides a storage medium which is applied to the server, and at least one program code is stored in the storage medium and is used for being executed by a processor and realizing the method for analyzing the emotion of the sentence in the embodiment of the application.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program for instructing relevant hardware, and the program may be stored in a storage medium, where the storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The foregoing is illustrative of the present application and is not to be construed as limiting thereof, but rather as various modifications, equivalent arrangements, improvements, etc., which fall within the spirit and principles of the present application.
Claims (18)
1. A method of emotion analysis for a sentence, the method comprising:
determining first word vectors corresponding to a plurality of words included in a sentence to be analyzed;
acquiring the dependency relationship among the plurality of words, wherein the dependency relationship is determined based on the part of speech of the words and the action of the words in sentences and is used for indicating that the words in the sentences are associated;
Determining a relation matrix according to the dependency relation, wherein elements in the relation matrix are used for representing whether the dependency relation exists among the vocabularies;
for any dependency, according to the type of the dependency, obtaining a corresponding relation vector of the dependency, wherein the type of the dependency comprises noun subjects, predicates, compound words, adjectives and objects of prepositions;
processing the relation matrix and the plurality of relation vectors based on a graph rolling network to obtain second word vectors corresponding to the plurality of words;
and determining the emotion type of at least one first target vocabulary in the plurality of vocabularies based on the second word vectors corresponding to the plurality of vocabularies.
2. The method of claim 1, wherein determining a first word vector corresponding to a plurality of words included in the sentence to be analyzed comprises:
word segmentation is carried out on sentences to be analyzed to obtain a plurality of words;
and for any vocabulary, splicing word vectors corresponding to the first meaning of the vocabulary with word vectors corresponding to the second meaning of the vocabulary to obtain the first word vectors corresponding to the vocabulary.
3. The method of claim 1, wherein the processing the relationship matrix and the plurality of relationship vectors based on the graph rolling network to obtain the second word vector corresponding to the plurality of words comprises:
For any vocabulary in the plurality of vocabularies, acquiring a first word vector corresponding to at least one vocabulary with a dependency relationship with the vocabulary;
and inputting the relation matrix, the plurality of relation vectors and the first word vector corresponding to the at least one vocabulary into the graph rolling network to obtain the second word vector corresponding to the vocabulary.
4. The method of claim 1, wherein determining the emotion type to which at least one first target vocabulary of the plurality of vocabularies belongs based on the second word vectors corresponding to the plurality of vocabularies comprises:
processing first word vectors corresponding to the plurality of words based on a first convolutional neural network to obtain third word vectors corresponding to the plurality of words, wherein the third word vectors are in one-to-one correspondence with the second word vectors;
splicing the corresponding second word vector and the third word vector to obtain a fourth word vector corresponding to the plurality of words;
and determining the emotion type of at least one first target vocabulary in the plurality of vocabularies based on the fourth word vector corresponding to the plurality of vocabularies.
5. The method of claim 1, wherein determining the emotion type to which at least one first target vocabulary of the plurality of vocabularies belongs based on the second word vectors corresponding to the plurality of vocabularies comprises:
Determining a first information vector corresponding to the plurality of words and a second information vector corresponding to the plurality of words according to a second word vector corresponding to the plurality of words, wherein the first information vector is used for indicating whether any word in the plurality of words is a first target word, and the second information vector is used for indicating the emotion type of each word in the plurality of words;
for each vocabulary, when the probability that the vocabulary belongs to the first target vocabulary is determined to be larger than the target probability threshold according to the first information vector corresponding to the vocabulary, the probability that the vocabulary belongs to each emotion type is determined according to the second information vector corresponding to the vocabulary.
6. The method of claim 5, wherein determining the first information vector corresponding to the plurality of words and the second information vector corresponding to the plurality of words from the second word vector corresponding to the plurality of words comprises:
recoding second word vectors corresponding to the plurality of words based on a second convolutional neural network to obtain first information vectors corresponding to the plurality of words;
recoding second word vectors corresponding to the plurality of words based on a third convolutional neural network to obtain second information vectors corresponding to the plurality of words;
Splicing the first information vector and the second information vector to obtain a third information vector;
and respectively recoding the third information vector based on the second convolutional neural network and the third convolutional neural network to obtain an updated first information vector and a second information vector.
7. The method of claim 6, wherein after the updated first information vector and second information vector are obtained, the method further comprises:
splicing the updated first information vector and the updated second information vector to obtain an updated third information vector;
re-encoding the updated third information vector based on the second convolutional neural network and the third convolutional neural network, respectively, again;
the above-described splicing and re-encoding process is repeated until a stop update condition is reached.
8. The method of claim 6, wherein the re-encoding the third information vector based on the second convolutional neural network and the third convolutional neural network, respectively, to obtain updated first information vector and second information vector, comprises:
the interactive transmission information obtained by the second convolutional neural network according to the second word vectors corresponding to the plurality of words is obtained, and the interactive transmission information is used for indicating words belonging to a second target word in the plurality of words;
Recoding the third information vector based on the third convolutional neural network and the interactive transfer information to obtain an updated second information vector;
and recoding the third information vector based on the second convolutional neural network to obtain an updated first information vector.
9. An apparatus for emotion analysis of a sentence, said apparatus comprising:
the determining module is used for determining first word vectors corresponding to a plurality of words included in the sentences to be analyzed;
the processing module is used for acquiring the dependency relationship among the plurality of words, wherein the dependency relationship is determined based on the part of speech of the words and the action of the words in sentences and is used for indicating that the words in the sentences are associated; determining a relation matrix according to the dependency relation, wherein elements in the relation matrix are used for representing whether the dependency relation exists among the vocabularies; for any dependency, according to the type of the dependency, obtaining a corresponding relation vector of the dependency, wherein the type of the dependency comprises noun subjects, predicates, compound words, adjectives and objects of prepositions; processing the relation matrix and the plurality of relation vectors based on a graph rolling network to obtain second word vectors corresponding to the plurality of words;
The determining module is further configured to determine, based on the second word vectors corresponding to the plurality of words, an emotion type to which at least one first target word in the plurality of words belongs.
10. The apparatus of claim 9, wherein the determining module is further configured to segment a sentence to be analyzed to obtain a plurality of words; and for any vocabulary, splicing word vectors corresponding to the first meaning of the vocabulary with word vectors corresponding to the second meaning of the vocabulary to obtain the first word vectors corresponding to the vocabulary.
11. The apparatus of claim 9, wherein the processing module is further configured to obtain, for any one of the plurality of words, a first word vector corresponding to at least one word having a dependency relationship with the word; and inputting the relation matrix, the plurality of relation vectors and the first word vector corresponding to the at least one vocabulary into the graph rolling network to obtain the second word vector corresponding to the vocabulary.
12. The apparatus of claim 9, wherein the determining module is further configured to process first word vectors corresponding to the plurality of words based on a first convolutional neural network to obtain third word vectors corresponding to the plurality of words, the third word vectors corresponding to the second word vectors one to one; splicing the corresponding second word vector and the third word vector to obtain a fourth word vector corresponding to the plurality of words; and determining the emotion type of at least one first target vocabulary in the plurality of vocabularies based on the fourth word vector corresponding to the plurality of vocabularies.
13. The apparatus of claim 9, wherein the determining module is further configured to determine a first information vector corresponding to the plurality of words and a second information vector corresponding to the plurality of words according to a second word vector corresponding to the plurality of words, the first information vector being used to indicate whether any word of the plurality of words is a first target word, the second information vector being used to indicate an emotion type to which each word of the plurality of words belongs; for each vocabulary, when the probability that the vocabulary belongs to the first target vocabulary is determined to be larger than the target probability threshold according to the first information vector corresponding to the vocabulary, the probability that the vocabulary belongs to each emotion type is determined according to the second information vector corresponding to the vocabulary.
14. The apparatus of claim 13, wherein the determining module is further configured to re-encode second word vectors corresponding to the plurality of words based on a second convolutional neural network to obtain first information vectors corresponding to the plurality of words; recoding second word vectors corresponding to the plurality of words based on a third convolutional neural network to obtain second information vectors corresponding to the plurality of words; splicing the first information vector and the second information vector to obtain a third information vector; and respectively recoding the third information vector based on the second convolutional neural network and the third convolutional neural network to obtain an updated first information vector and a second information vector.
15. The apparatus of claim 14, wherein the apparatus further comprises:
the splicing module is used for splicing the updated first information vector and the updated second information vector to obtain an updated third information vector;
the coding module is used for recoding the updated third information vector again based on the second convolutional neural network and the third convolutional neural network respectively;
and the updating module is used for repeating the splicing and recoding processes until the updating stopping condition is reached.
16. The apparatus of claim 14, wherein the determining module is further configured to obtain interaction transfer information obtained by the second convolutional neural network according to a second word vector corresponding to the plurality of words, where the interaction transfer information is used to indicate a word belonging to a second target word in the plurality of words; recoding the third information vector based on the third convolutional neural network and the interactive transfer information to obtain an updated second information vector; and recoding the third information vector based on the second convolutional neural network to obtain an updated first information vector.
17. A server comprising a processor and a memory for storing at least one piece of program code, the at least one piece of program code being loaded by the processor and performing the method of semantic emotion analysis of any of claims 1 to 8.
18. A storage medium storing at least one piece of program code for execution by a processor and for performing the method of emotion analysis of a sentence according to any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911147402.3A CN110929528B (en) | 2019-11-21 | 2019-11-21 | Method, device, server and storage medium for analyzing emotion of sentence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911147402.3A CN110929528B (en) | 2019-11-21 | 2019-11-21 | Method, device, server and storage medium for analyzing emotion of sentence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110929528A CN110929528A (en) | 2020-03-27 |
| CN110929528B true CN110929528B (en) | 2023-09-05 |
Family
ID=69851465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911147402.3A Active CN110929528B (en) | 2019-11-21 | 2019-11-21 | Method, device, server and storage medium for analyzing emotion of sentence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110929528B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111861546A (en) * | 2020-06-28 | 2020-10-30 | 浪潮(北京)电子信息产业有限公司 | Comment quality judgment method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103365867A (en) * | 2012-03-29 | 2013-10-23 | 腾讯科技（深圳）有限公司 | Method and device for emotion analysis of user evaluation |
| CN103455562A (en) * | 2013-08-13 | 2013-12-18 | 西安建筑科技大学 | Text orientation analysis method and product review orientation discriminator on basis of same |
| CN110377740A (en) * | 2019-07-22 | 2019-10-25 | 腾讯科技（深圳）有限公司 | Feeling polarities analysis method, device, electronic equipment and storage medium |
| WO2019214145A1 (en) * | 2018-05-10 | 2019-11-14 | 平安科技（深圳）有限公司 | Text sentiment analyzing method, apparatus and storage medium |
| CN110472003A (en) * | 2019-08-08 | 2019-11-19 | 东北大学 | Social networks text emotion fine grit classification method based on figure convolutional network |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140365206A1 (en) * | 2013-06-06 | 2014-12-11 | Xerox Corporation | Method and system for idea spotting in idea-generating social media platforms |
| US10089296B2 (en) * | 2015-12-30 | 2018-10-02 | Nice Ltd. | System and method for sentiment lexicon expansion |
-
2019
- 2019-11-21 CN CN201911147402.3A patent/CN110929528B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103365867A (en) * | 2012-03-29 | 2013-10-23 | 腾讯科技（深圳）有限公司 | Method and device for emotion analysis of user evaluation |
| CN103455562A (en) * | 2013-08-13 | 2013-12-18 | 西安建筑科技大学 | Text orientation analysis method and product review orientation discriminator on basis of same |
| WO2019214145A1 (en) * | 2018-05-10 | 2019-11-14 | 平安科技（深圳）有限公司 | Text sentiment analyzing method, apparatus and storage medium |
| CN110377740A (en) * | 2019-07-22 | 2019-10-25 | 腾讯科技（深圳）有限公司 | Feeling polarities analysis method, device, electronic equipment and storage medium |
| CN110472003A (en) * | 2019-08-08 | 2019-11-19 | 东北大学 | Social networks text emotion fine grit classification method based on figure convolutional network |
Non-Patent Citations (1)
| Title |
|---|
| 利用动态多池卷积神经网络的情感分析模型;喻涛;罗可;;计算机科学与探索(第07期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110929528A (en) | 2020-03-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11334635B2 (en) | Domain specific natural language understanding of customer intent in self-help | |
| CN105989040B (en) | Intelligent question and answer method, device and system | |
| US20180082184A1 (en) | Context-aware chatbot system and method | |
| US20210390609A1 (en) | System and method for e-commerce recommendations | |
| US10360219B2 (en) | Applying level of permanence to statements to influence confidence ranking | |
| US20160328403A1 (en) | Method and system for app search engine leveraging user reviews | |
| US10032448B1 (en) | Domain terminology expansion by sensitivity | |
| US10825071B2 (en) | Adaptive multi-perceptual similarity detection and resolution | |
| US11144569B2 (en) | Operations to transform dataset to intent | |
| US11874798B2 (en) | Smart dataset collection system | |
| CN110019948B (en) | Method and apparatus for outputting information | |
| US20170286970A1 (en) | Answer-suggestion system for automatically resolving customer requests | |
| CN111931055A (en) | Object recommendation method, object recommendation device and electronic equipment | |
| CN108304381B (en) | Entity edge establishing method, device and equipment based on artificial intelligence and storage medium | |
| CN110929528B (en) | Method, device, server and storage medium for analyzing emotion of sentence | |
| CN114116997A (en) | Knowledge question answering method, knowledge question answering device, electronic equipment and storage medium | |
| CN116662495A (en) | Question-answering processing method, and method and device for training question-answering processing model | |
| KR102601804B1 (en) | Electronic device and method for providing recruitment recommendation service based on neural network model | |
| CN112328799B (en) | Question classification method and device | |
| CN113094584A (en) | Method and device for determining recommended learning resources | |
| CN114090725A (en) | Emotion prediction model training method and device | |
| CN113239284B (en) | Method and device for determining sequence of recommendation information | |
| CN110598040B (en) | Album recall method, device, equipment and storage medium | |
| US20240046042A1 (en) | Method and device for information processing | |
| Kuzmin | Item2vec-based approach to a recommender system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40021093 Country of ref document: HK |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |