CN110969011B - Text emotion analysis method and device, storage medium and processor - Google Patents
Text emotion analysis method and device, storage medium and processor Download PDFInfo
- Publication number
- CN110969011B CN110969011B CN201811159908.1A CN201811159908A CN110969011B CN 110969011 B CN110969011 B CN 110969011B CN 201811159908 A CN201811159908 A CN 201811159908A CN 110969011 B CN110969011 B CN 110969011B
- Authority
- CN
- China
- Prior art keywords
- target
- text
- analyzed
- words
- sentence
- 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
- 230000008451 emotion Effects 0.000 title claims abstract description 228
- 238000004458 analytical method Methods 0.000 title claims abstract description 70
- 230000000007 visual effect Effects 0.000 claims abstract description 101
- 238000007619 statistical method Methods 0.000 claims abstract description 22
- 230000007935 neutral effect Effects 0.000 claims description 52
- 238000000034 method Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 8
- 230000011218 segmentation Effects 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 5
- 238000013145 classification model Methods 0.000 abstract description 11
- 230000015654 memory Effects 0.000 description 28
- 238000004590 computer program Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000002860 competitive effect Effects 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- 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
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a text sentiment analysis method, a text sentiment analysis device, a storage medium and a processor, wherein target visual angle words are identified in a text to be analyzed, the text to be analyzed is subjected to statistical analysis according to the target visual angle words, the sentiment category of the text to be analyzed at a target visual angle is preliminarily judged, and/or the sentiment category of a sentence in the text to be analyzed at the target visual angle is preliminarily judged, then the sentiment classification of the sentence at the target visual angle is obtained by calculating the target sentence in the sentences of which the sentiment category at the target visual angle is not determined through a sentence sentiment classification model, and finally the sentiment classification results of the sentences at the target visual angle are combined to obtain the final visual angle sentiment classification of the text to be analyzed at the target visual angle. The emotion analysis of the text at different viewing angles is realized, the problem that judgment of different viewing angles is influenced mutually can be avoided by analyzing the emotion analysis of the text at a single viewing angle, and the emotion expressed by the text aiming at the single viewing angle can be determined more accurately.
Description
Technical Field
The present invention relates to the field of information processing technologies, and in particular, to a text emotion analysis method, apparatus, storage medium, and processor.
Background
Emotion analysis has become a hot issue in the field of natural speech processing in recent years. With the development of forums, microblogs, wechat and other communication media, more and more speech information is used to express the view or preference of users to things. For merchants or manufacturers, it is particularly important to perform emotion analysis on related products and competitive products from massive network information. And the data volume is explosively increased, and the emotion of the user on related products and competitive products is identified from the text of the long-text debate by manpower, which obviously cannot meet the requirements of the big data era.
However, most of the current emotion analysis is emotion analysis at sentence or phrase level, and most of the emotion analysis is attribute emotion analysis at fine granularity, but the related research on the perspective emotion analysis of texts at chapter level is very little, and at present, the emotion analysis on the texts at chapter level does not distinguish the perspective, and the whole article has only one emotion analysis result. In fact, the emotion presented in one article is different when standing at different viewing angles.
Therefore, how to analyze emotions from different perspectives for texts at chapter level becomes an urgent problem to be solved.
Disclosure of Invention
In view of the above, the present invention has been made to provide a data processing method, apparatus, storage medium and processor that overcome or at least partially solve the above problems.
In one aspect, the present application provides a text sentiment analysis method, including:
identifying target visual angle words used for representing target visual angles to be analyzed in texts to be analyzed;
performing statistical analysis on the text to be analyzed according to the target view angle words to judge whether the emotion types of the whole text to be analyzed at the target view angle are neutral, and/or judging whether the emotion types of sentences containing the target view angle words and/or expression words of the target view angle words in the text to be analyzed at the target view angle are neutral;
if at least one judgment result is negative, determining a target sentence in the sentences which are not determined to be in the emotion type of the target visual angle in the text to be analyzed, and extracting a feature sequence from the target sentence;
inputting the characteristic sequence into a pre-trained sentence emotion analysis model to obtain the emotion type of the target sentence at the target view angle;
and combining the emotion types of all the sentences determining the emotion types at the target view angle to obtain the emotion types of the text to be analyzed at the target view angle.
In another aspect, the present application provides a text emotion analyzing apparatus, including:
the identification module is used for identifying target visual angle words used for representing target visual angles to be analyzed in the text to be analyzed;
the analysis module is used for carrying out statistical analysis on the text to be analyzed according to the target view angle words so as to judge whether the emotion types of the whole text to be analyzed at the target view angle are neutral or not, and/or judge whether the emotion types of sentences containing the target view angle words and/or expression words of the target view angle words in the text to be analyzed at the target view angle are neutral or not;
the extraction module is used for determining a target sentence in the sentence which is not determined to be the emotion type of the target view angle in the text to be analyzed and extracting a feature sequence from the target sentence if at least one judgment result is negative;
the classification module is used for inputting the characteristic sequence into a pre-trained sentence emotion analysis model to obtain the emotion type of the target sentence at the target view angle;
and the merging module is used for merging the emotion types of the sentences of which the emotion types are determined at the target view angle to obtain the emotion types of the text to be analyzed at the target view angle.
In a third aspect, the present application provides a storage medium, where the storage medium includes a stored program, and when the program runs, a device in which the storage medium is located is controlled to execute the text emotion analysis method as described above.
In a fourth aspect, the present application provides a processor for executing a program, where the program executes to perform the text emotion analysis method as described above.
By means of the technical scheme, the text sentiment analysis method, the text sentiment analysis device, the storage medium and the processor provided by the invention have the advantages that target view angle words are identified in a text to be analyzed, the text to be analyzed is subjected to statistical analysis according to the target view angle words, the sentiment category of the text to be analyzed at a target view angle is preliminarily judged, and/or the sentiment category of a sentence in the text to be analyzed at the target view angle is preliminarily judged, then the target sentence in the sentence which is not determined to be the sentiment category at the target view angle is calculated through a sentence sentiment classification model to obtain the sentiment classification of the sentence at the target view angle, and finally the sentiment classification results of the sentence at the target view angle are combined to obtain the final view angle sentiment classification of the text to be analyzed at the target view angle. The emotion analysis of the text at different viewing angles is realized, the problem that judgment of different viewing angles is influenced mutually can be avoided by analyzing the emotion analysis of the text at a single viewing angle, and the emotion expressed by the text aiming at the single viewing angle can be determined more accurately.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow chart of one implementation of a text emotion analysis method provided by an embodiment of the present application;
FIG. 2 is a flow chart of one implementation of extracting a feature sequence in a target sentence according to an embodiment of the present application;
fig. 3 shows a schematic structural diagram of a text emotion analysis device provided in an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
An implementation flowchart of the text emotion analysis method provided in the embodiment of the present application is shown in fig. 1, and may include:
step S11: and identifying target view angle words used for representing the target view angle to be analyzed in the text to be analyzed.
The target perspective word is a word in the text to be analyzed, and the word represents a perspective.
Taking the text in the automobile field as an example, the target view angle to be analyzed may be a certain vehicle type, such as a tourist, a bmw, etc. Correspondingly, if the target view is a view, the corresponding target view word may be: tuo, alternatively, tigian; if the target view is a BMW, the corresponding target view words are: BMW, or BMW.
If the emotion classification of the text to be analyzed at the perspective of the enroute is to be analyzed, the enroute, or Tiguan, can be identified from the text to be analyzed; if the emotion classification of the text to be analyzed in the perspective of the BMW is analyzed, the BMW or the BMW can be identified in the text to be analyzed.
In the embodiment of the present application, a sequence labeling method of a Conditional Random Field (CRF) model may be used to identify target perspective words in a text to be analyzed. Or, the perspective word dictionary may be used to perform perspective word recognition, that is, the perspective words in the dictionary are directly used to match words in the text. For convenience of subsequent calculation, in addition to identifying the target perspective word, an expression word of the target perspective word can be identified.
The target perspective words and the expression words of the target perspective words are different description modes of the same thing, and the expression words of the target perspective words can refer to short names or alternative names of the target perspective words. For example, if the target perspective word is "look-in-the-way", the expression word for "look-in-the-way" is tigean, and if the target perspective word is "BMW", the expression word for "BMW".
It should be noted that if only emotion analysis is performed on a text to be recognized at a certain viewing angle, only one target viewing angle word may be recognized, and if emotion analysis is required to be performed on the text to be recognized at N (N is a positive integer greater than 1) viewing angles, N target viewing angle words need to be recognized in the text to be recognized, and different target viewing angle words represent different target viewing angles.
In an optional embodiment, the text to be analyzed refers to a text in a certain field screened from a large amount of data, such as a text related to an automobile field, and the text in the automobile field may be obtained by performing text classification on the large amount of data through a Support Vector Machine (SVM) classification model, a Logistic Regression (LR) classification model, a Long Short-Term Memory network (LSTM) classification model, or the like.
After obtaining the text in the automobile field, in order to improve the processing efficiency of the emotion analysis of the text, the classified data (e.g., advertisement, car selling, maintenance, modification, etc.) irrelevant to emotion in the text in the automobile field may be filtered, for example, words such as "remuneration," maintenance skill, "and" time-limited special sale "are filtered. And filtering to obtain the text to be analyzed.
Step S12: and performing statistical analysis on the text to be analyzed according to the target view angle words to judge whether the emotion types of the whole text to be analyzed at the target view angle are neutral, and/or judging whether the emotion types of sentences containing the target view angle words and/or expression words of the target view angle words in the text to be analyzed at the target view angle are neutral.
Wherein, performing statistical analysis on the text to be analyzed may include: and performing statistical analysis on the text to be analyzed according to the number of sentences containing the target visual angle in the text to be analyzed, the number of visual angles contained in the sentences containing the target visual angle in the text to be analyzed, and the position of the target visual angle in the text to be analyzed. The emotion type of the whole text to be analyzed at the target view angle is judged to be neutral, and the emotion type of a sentence containing the target view angle words and/or expression words of the target view angle words in the text to be analyzed at the target view angle is judged to be neutral.
The emotion categories may include the following three: positive, negative and neutral.
In an alternative embodiment, the emotion classification of the whole text to be analyzed in the target view angle may be determined first. If the emotion classification of the whole text to be analyzed in the target view angle is judged to be neutral, the subsequent process is not needed. If the emotion classification of the whole text to be analyzed at the target view angle is judged to be not neutral, whether the emotion classification of a sentence, which contains target view angle words and/or expression words of the target view angle words, in the text to be analyzed at the target view angle is neutral or not can be judged, and if the emotion classification of a sentence at the target view angle is judged to be neutral, the emotion classification of the sentence does not need to be calculated subsequently.
In another optional embodiment, it may not be necessary to determine the emotion classification of the whole text to be analyzed at the target view angle, but it may be directly determined whether the emotion classification of a sentence in the text to be analyzed, which includes the target view angle word and/or the expression word of the target view angle word, at the target view angle is neutral, and if it is determined that the emotion classification of a sentence at the target view angle is neutral, it may not be necessary to calculate the emotion classification of the sentence subsequently.
Step S13: and if at least one judgment result is negative, determining a target sentence in the sentences which are not determined to be in the emotion types of the target visual angle in the text to be analyzed, and extracting a characteristic sequence from the target sentence.
The sentence of the emotion category not determined in the target view may or may not include the target view word or the expression word of the target view word. Accordingly, the determined target sentence may or may not include the target perspective word or the expression word of the target perspective word.
Step S14: and inputting the characteristic sequence into a pre-trained sentence emotion analysis model to obtain the emotion type of the target sentence at the target view angle.
In the embodiment of the application, for the target sentences of which the emotion types at the target view angles are not determined through statistical analysis, the emotion types of the target sentences are calculated by using the sentence emotion analysis model.
In an alternative embodiment, the sentence emotion analysis model may be a neural network model, and the neural network model may be: a long-short term memory network model based on attention mechanism.
Step S15: and combining the emotion types of all the sentences determining the emotion types at the target view angle to obtain the emotion types of the text to be analyzed at the target view angle.
When emotion category merging is performed, the emotion category of the sentence in the target view can be assigned, for example, the positive assignment is 1, the negative assignment is-1, and the neutral assignment is 0. Then, merging the emotion classifications of all the sentences for which emotion classifications are determined at the target view may include:
summing the values of the emotion categories of the sentences determining the emotion categories at the target view angle to obtain the emotion category value of the text to be analyzed at the target view angle, wherein if the emotion category value is greater than 0, the emotion categories to be analyzed at the target view angle are positive, if the emotion category value is less than 0, the emotion categories to be analyzed at the target view angle are negative, and if the emotion category value is equal to 0, the emotion categories to be analyzed at the target view angle are neutral.
According to the text sentiment analysis method, target visual angle words are identified in a text to be analyzed, the text to be analyzed is subjected to statistical analysis according to the target visual angle words, the sentiment category of the text to be analyzed at a target visual angle is preliminarily judged, and/or the sentiment category of a sentence in the text to be analyzed at the target visual angle is preliminarily judged, then the target sentence in the sentence which is not determined to be in the sentiment category at the target visual angle is calculated through a sentence sentiment classification model to obtain the sentiment classification of the sentence at the target visual angle, and finally the sentiment classification results of the sentence at the target visual angle are combined to obtain the final visual angle sentiment classification of the text to be analyzed at the target visual angle. The emotion analysis of the text at different viewing angles is realized, the problem that judgment of different viewing angles is influenced mutually can be avoided by analyzing the emotion analysis of the text at a single viewing angle, and the emotion expressed by the text aiming at the single viewing angle can be determined more accurately.
In addition, the calculation amount of statistical analysis is far less than that of calculating the emotion type of the text and/or the emotion type of the sentence based on the sentence emotion classification model, so compared with the case that the emotion type of the sentence and the emotion type of the text are calculated through the emotion classification model, the text emotion analysis method provided by the application is less in calculation amount and higher in analysis efficiency. By which the classification efficiency of the scheme of the present application can be higher when emotion analysis needs to be performed from multiple perspectives, respectively.
In an alternative embodiment, the flowchart of one implementation of extracting a feature sequence in a target sentence is shown in fig. 2, and may include:
step S21: and performing word segmentation processing on the target sentence to obtain a plurality of word segments. Stop words are not included in the plurality of segmented words.
For example, assume that the target statement is: the oil consumption of the A vehicle type is lower than that of the B vehicle type. The result of performing word segmentation on the target sentence is as follows: the vehicle type A is lower in oil consumption than the vehicle type B, wherein stop words are filtered out.
Step S22: constructing a feature sequence based on the plurality of word segments, wherein the feature sequence comprises: each participle, part of speech corresponding to each participle, identification corresponding to each participle, statement word of target view word, and position of each participle relative to the target view word or statement word of the target view word.
Wherein, the above-mentioned sign includes: view identification, attribute identification, rating identification, or other identification. The visual angle identification is used for distinguishing different visual angles, the attribute identification is used for distinguishing different attributes, the evaluation identification is used for distinguishing different evaluations, and the other identifications are identifications except for the visual angle identification, the attribute identification and the evaluation identification.
Based on the above example, assuming that the target view angle is the B vehicle type, the obtained feature sequence is: the fuel consumption of the vehicle A is lower than that of the vehicle B by a noun adjective first visual angle identification attribute identification and other identifications of a second visual angle identification evaluation identification, and the identification is not-3-2-1. Here, "none" means that the sentence does not include the expression of "B vehicle type".
It should be noted that, if the target statement does not include a certain item in the feature sequence, the value of the item in the feature sequence is filled with a preset value. For example, in the above example, if the sentence does not include the expression word of the "B vehicle type", the value of the term "expression word of the target perspective word" in the feature sequence is the value corresponding to "none", which is, of course, only illustrated here, and may be other values.
In addition, if the target sentence includes both the target perspective word and the expression word of the target perspective word, or the target sentence includes two expression words of the target perspective word, it is necessary to extract the feature sequences corresponding to the two expression words, respectively, and since the positions of the two expression words in the sentence are different, the two extracted feature sequences are different in the value of the term "the position of each participle with respect to the target perspective word or the expression word of the target perspective word". Subsequently, when calculating the emotion classification of the target sentence at the target view angle, the emotion classification of the target sentence at the target view angle needs to be calculated according to the two extracted feature sequences, and then the obtained two emotion classifications of the target sentence at the target view angle are combined to obtain the final emotion classification of the target sentence at the target view angle.
The specific step of classifying and combining the two emotions of the target sentence at the target view angle may be: summing the values of the two emotion classifications of the target sentence at the target view angle to obtain a value of a final emotion classification of the target sentence at the target view angle, if the value of the final emotion classification is less than 0, the final emotion classification of the target sentence at the target view angle is negative, if the value of the final emotion classification is more than 0, the final emotion classification of the target sentence at the target view angle is positive, and if the value of the final emotion classification is equal to 0, the final emotion classification of the target sentence at the target view angle is neutral.
In this embodiment, when the emotion classification of a sentence is calculated through the sentence training model, the relative position information of the participle with respect to the target view-angle word or the expression word of the target view-angle word and the related information of the view-angle word (e.g., the expression word of the target view-angle word, or the view-angle identifier, etc.) are added to the feature sequence, so that the accuracy of emotion classification of the sentence at the target view angle is improved.
In an optional embodiment, according to the target perspective word, one implementation manner of performing statistical analysis on the text to be analyzed may be:
if the title of the text to be analyzed does not contain the target visual angle words and the expression words of the target visual angle words, the number of sentences of the text to be analyzed is greater than a first preset threshold, and the number of the sentences containing the target visual angle words and/or the expression words of the target visual angle words in the text is less than a second preset threshold, determining that the emotion type of the whole text to be analyzed at the target visual angle is neutral.
Generally, the text to be analyzed includes two parts, a title and a body. In this embodiment, if the title of the text to be analyzed does not include the target perspective (that is, the title of the text to be analyzed does not include the target perspective word and the expression word of the target perspective word), the number of the sentences included in the body is greater than the first preset threshold (for example, 30), and only a small number of the sentences in the body include the target perspective (for example, only 1 sentence includes the target perspective), the emotion category of the whole text to be analyzed in the target perspective is considered to be neutral.
If the emotion type of the whole text to be analyzed at the target visual angle is not neutral, counting the number of visual angles contained in a first sentence containing the target visual angle in the text to be analyzed; and if the number of the visual angles is larger than a third preset threshold (for example, 5), determining that the emotion type of the first sentence at the target visual angle is neutral.
The first sentence is any sentence in the text to be analyzed. The emotion classification of the first sentence in each view included in the first sentence is neutral.
In the above embodiment, the emotion type of the whole text to be analyzed at the target view angle is determined, and when the emotion type of the text to be analyzed at the target view angle is not neutral, whether the emotion type of the sentence at the target view angle is neutral is determined. In another optional embodiment, the emotion classification of the whole text to be analyzed in the target view angle may not be determined, but whether the emotion classification of each sentence in the text to be analyzed in the target view angle is neutral may be directly determined.
In this embodiment, the emotion type of the text or sentence to be analyzed at the target view angle is preliminarily analyzed according to the preset statistical analysis rule, and it is obtained that the emotion type of the text or sentence to be analyzed at the target view angle is neutral or not. The method improves the accuracy of emotion classification of the text to be analyzed at the target view angle.
In an optional embodiment, one implementation manner of determining the target sentence in the sentence of the emotion category not determined at the target view angle in the text to be analyzed may be:
if the text to be analyzed only contains the target perspective words and/or the expression words of the target perspective words, and the number of the sentences containing the expression words of the target perspective words and/or the target perspective words is smaller than a fourth preset threshold, then if the sentences containing the expression words of the target perspective words and/or the target perspective words include the titles of the text to be analyzed, and the number of the sentences in the text of the text to be analyzed is smaller than a fifth preset threshold, all the sentences which are not determined to be in the emotion categories at the target perspective in the text of the text to be analyzed are taken as the target sentences.
In this embodiment, under the condition that the text to be analyzed only includes the target view angle and only a small number of sentences include the target view angle, if the sentences including the target view angle include titles and the number of the sentences of the text to be analyzed is less than a fifth preset value, it is considered that the whole emotion of the text to be analyzed affects the emotion of the target view angle, and at this time, all the sentences not determined as the emotion types at the target view angle in the text body of the text to be analyzed are taken as the target sentences.
For example, if the text to be analyzed only includes the target view angle and only one sentence includes the target view angle, if the sentence including the target view angle is the title and the number of sentences of the text to be analyzed is less than 5, it is considered that all sentences in the text to be analyzed should calculate their emotion types at the target view angle, and therefore all sentences not determined as emotion types at the target view angle are taken as target sentences.
In this embodiment, except for calculating the emotion category of the sentence including the target view angle at the target view angle, the emotion category of the sentence not including the target view angle at the target view angle is also calculated, so that the text emotion classification method provided by the present application has a higher classification accuracy.
In an optional embodiment, another implementation manner of determining the target sentence in the sentences of which the emotion category in the target view is not determined in the text to be analyzed may be:
if the text to be analyzed only contains the target perspective words and/or the expression words of the target perspective words, and the number of the sentences containing the expression words of the target perspective words and/or the target perspective words is smaller than a fourth preset threshold, then if the sentences containing the expression words of the target perspective words and/or the target perspective words do not include the title of the text to be analyzed, and the number of the sentences in the text of the text to be analyzed is smaller than a sixth preset threshold, the sentences which are not determined to be in the target perspective and have a length larger than the seventh preset threshold in the text to be analyzed are taken as the target sentences.
In this embodiment, when the text to be analyzed only includes the target view angle and only a small number of sentences include the target view angle, if the title does not include the target view angle and the number of sentences of the text to be analyzed is less than a sixth preset value (e.g., 4), it is considered that the emotion of a long sentence in the text to be analyzed affects the emotion of the target view angle, and at this time, a sentence which is not determined as the emotion category at the target view angle and has a length greater than a seventh preset threshold value (e.g., 3) in the text of the text to be analyzed is taken as the target sentence.
In the text emotion classification method, besides calculating the emotion classification of the sentence including the target view angle at the target view angle, the emotion classification of the sentence not including the target view angle at the target view angle may also be calculated.
Further, an implementation manner of taking the sentence with the length greater than the seventh preset threshold in the text to be analyzed as the target sentence may be:
taking sentences of which the lengths are greater than a seventh preset threshold value and less than an eighth preset threshold value (such as 60) in the texts to be analyzed as target sentences; the eighth preset threshold is greater than the seventh preset threshold.
In the present application, in order to avoid that the length of a sentence is too long due to the absence of punctuation marks, and thus the emotion classification accuracy of a text is reduced, in this embodiment, only a sentence with a length within a certain range is selected as a target sentence.
Corresponding to the embodiment of the method, the present application further provides a text emotion analysis device, and a schematic structural diagram of the text emotion analysis device provided by the present application is shown in fig. 3, and may include:
a recognition module 31, an analysis module 32, an extraction module 33, a classification module 34 and a merging module 35; wherein,
the identification module 31 is configured to identify a target perspective word used for representing a target perspective to be analyzed in a text to be analyzed;
the analysis module 32 is configured to perform statistical analysis on the text to be analyzed according to the target view word to determine whether the emotion category of the whole text to be analyzed at the target view is neutral, and/or determine whether the emotion category of a sentence in the text to be analyzed, which includes the target view word and/or an expression word of the target view word, at the target view is neutral;
the extraction module 33 is configured to determine a target sentence in the sentence, which is not determined to be the emotion category of the target view, in the text to be analyzed if at least one of the determination results is no, and extract a feature sequence in the target sentence;
the classification module 34 is configured to input the feature sequence into a pre-trained sentence emotion analysis model to obtain an emotion category of the target sentence at the target view;
the merging module 35 is configured to merge emotion categories of all sentences for determining emotion categories at the target view angle to obtain emotion categories of the text to be analyzed at the target view angle.
The text sentiment analysis device identifies target visual angle words in a text to be analyzed, firstly carries out statistical analysis on the text to be analyzed according to the target visual angle words, carries out preliminary judgment on the sentiment category of the text to be analyzed at a target visual angle, and/or carries out preliminary judgment on the sentiment category of a sentence in the text to be analyzed at the target visual angle, then calculates the sentiment classification of the sentence at the target visual angle according to a sentence sentiment classification model on the target sentence which is not determined to be in the sentiment category of the target visual angle, and finally combines the sentiment classification results of the sentences at the target visual angle to obtain the final visual angle sentiment classification of the text to be analyzed at the target visual angle. The emotion analysis of the text at different viewing angles is realized, the problem that judgment of different viewing angles is influenced mutually can be avoided by analyzing the emotion analysis of the text at a single viewing angle, and the emotion expressed by the text aiming at the single viewing angle can be determined more accurately.
In addition, the calculation amount of the statistical analysis is far less than the calculation amount of calculating the emotion type of the text and/or the emotion type of the sentence based on the sentence emotion classification model, so compared with the case that the emotion type of the sentence and the emotion type of the text are calculated by the emotion classification model, the text emotion analysis device provided by the application has the advantages of less calculation amount and higher analysis efficiency. By this, when emotion analysis needs to be performed from multiple viewing angles, the classification efficiency of the scheme of the application is higher.
In an optional embodiment, when the feature sequence in the target sentence is extracted by the extraction module 33, specifically:
performing word segmentation processing on the target sentence to obtain a plurality of words;
constructing a feature sequence based on the plurality of word segments, wherein the feature sequence comprises: each participle, part of speech corresponding to each participle, identification corresponding to each participle, a predicate of a target perspective word, and a position of each participle relative to the target perspective word or the predicate of the target perspective word;
the identification comprises: view identification, attribute identification, rating identification, or other identification.
In an alternative embodiment, the analysis module 32 may be specifically configured to:
if the title of the text to be analyzed does not contain the target visual angle words and the expression words of the target visual angle words, the number of sentences of the text to be analyzed is greater than a first preset threshold value, and the number of sentences containing the target visual angle words and/or the expression words of the target visual angle words in the text is less than a second preset threshold value, determining that the emotion type of the whole text to be analyzed at the target visual angle is neutral;
if the emotion type of the whole text to be analyzed in the target view angle is not neutral, counting the number of view angles contained in a first sentence containing the target view angle in the text to be analyzed; and if the number of the visual angles is larger than a third preset threshold value, determining that the emotion type of the first sentence at the target visual angle is neutral.
In an optional embodiment, when determining the target sentence, the extracting module 33 is specifically configured to:
if the text to be analyzed only includes the target perspective words and/or the expression words of the target perspective words, and the number of the sentences including the expression words of the target perspective words and/or the target perspective words is smaller than a fourth preset threshold, if the sentences including the expression words of the target perspective words and/or the target perspective words include the titles of the text to be analyzed, and the number of the sentences of the text to be analyzed is smaller than a fifth preset threshold, all the sentences which are not determined to be in the target perspective in the text of the text to be analyzed are taken as target sentences.
Or,
if the text to be analyzed only includes the target perspective word and/or the expression word of the target perspective word, and the number of the sentences including the target perspective word and/or the expression word of the target perspective word is less than a fourth preset threshold, then, if the sentences including the expression word of the target perspective word and/or the target perspective word do not include the title of the text to be analyzed, and the number of the sentences of the text to be analyzed is less than a sixth preset threshold, the sentences which are not determined to be in the target perspective and have a length greater than a seventh preset threshold in the text to be analyzed are taken as the target sentences.
The data processing device comprises a processor and a memory, wherein the identification module, the analysis module, the extraction module, the classification module, the combination module and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more than one, emotion analysis is carried out on the text by adjusting kernel parameters, so that emotion analysis of the text at different viewing angles is realized, the problem that judgment of different viewing angles is influenced mutually is avoided, and emotion expressed by the text aiming at a single viewing angle can be determined more accurately.
The memory may include volatile memory in a computer readable medium, random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
An embodiment of the present invention provides a storage medium, on which a program is stored, and the program implements the text sentiment analysis method when executed by a processor.
The embodiment of the invention provides a processor, which is used for running a program, wherein the text emotion analysis method is executed when the program runs.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps:
a text emotion analysis method comprises the following steps:
identifying target visual angle words used for representing target visual angles to be analyzed in the text to be analyzed;
performing statistical analysis on the text to be analyzed according to the target view angle words to judge whether the emotion types of the whole text to be analyzed at the target view angle are neutral, and/or judge whether the emotion types of sentences containing the target view angle words and/or expression words of the target view angle words in the text to be analyzed at the target view angle are neutral;
if at least one judgment result is negative, determining a target sentence in the sentence which is not determined to be in the emotion type of the target visual angle in the text to be analyzed, and extracting a characteristic sequence from the target sentence;
inputting the characteristic sequence into a pre-trained sentence emotion analysis model to obtain the emotion type of the target sentence at the target view angle;
and combining the emotion types of all the sentences determining the emotion types at the target view angle to obtain the emotion types of the text to be analyzed at the target view angle.
Wherein the extracting the feature sequence in the target sentence comprises:
performing word segmentation processing on the target sentence to obtain a plurality of words;
constructing a feature sequence based on the plurality of word segments, wherein the feature sequence comprises: each participle, part of speech corresponding to each participle, identification corresponding to each participle, a predicate of a target perspective word, and a position of each participle relative to the target perspective word or the predicate of the target perspective word;
the identification comprises: view identification, attribute identification, rating identification, or other identification.
Wherein, the performing statistical analysis on the text to be analyzed according to the target perspective words comprises:
if the title of the text to be analyzed does not contain the target visual angle words and the expression words of the target visual angle words, the number of sentences of the text to be analyzed is greater than a first preset threshold value, and the number of sentences containing the target visual angle words and/or the expression words of the target visual angle words in the text is less than a second preset threshold value, determining that the emotion type of the whole text to be analyzed at the target visual angle is neutral;
if the emotion type of the whole text to be analyzed in the target view angle is not neutral, counting the number of view angles contained in a first sentence containing the target view angle in the text to be analyzed; and if the number of the visual angles is larger than a third preset threshold value, determining that the emotion type of the first sentence at the target visual angle is neutral.
Wherein, the determining a target sentence in the sentence of which the emotion category in the target view angle is not determined in the text to be analyzed comprises:
if the text to be analyzed only includes the target perspective words and/or the expression words of the target perspective words, and the number of the sentences including the expression words of the target perspective words and/or the target perspective words is smaller than a fourth preset threshold, if the sentences including the expression words of the target perspective words and/or the target perspective words include the titles of the text to be analyzed, and the number of the sentences of the text to be analyzed is smaller than a fifth preset threshold, all the sentences which are not determined to be in the target perspective in the text of the text to be analyzed are taken as target sentences.
Wherein, the determining a target sentence in the sentence of which the emotion category in the target view angle is not determined in the text to be analyzed comprises:
if the text to be analyzed only includes the target perspective word and/or the expression word of the target perspective word, and the number of the sentences including the target perspective word and/or the expression word of the target perspective word is smaller than a fourth preset threshold, then if the sentences including the target perspective word and/or the expression word of the target perspective word do not include the title of the text to be analyzed, and the number of the sentences of the text body of the text to be analyzed is smaller than a sixth preset threshold, the sentences which are not determined to be in the target perspective and have lengths larger than a seventh preset threshold in the text to be analyzed are taken as the target sentences.
The device herein may be a server, a PC, a PAD, a mobile phone, etc.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device:
a text sentiment analysis method comprises the following steps:
identifying target visual angle words used for representing target visual angles to be analyzed in the text to be analyzed;
performing statistical analysis on the text to be analyzed according to the target view angle words to judge whether the emotion types of the whole text to be analyzed at the target view angle are neutral, and/or judge whether the emotion types of sentences containing the target view angle words and/or expression words of the target view angle words in the text to be analyzed at the target view angle are neutral;
if at least one judgment result is negative, determining a target sentence in the sentences which are not determined to be in the emotion type of the target visual angle in the text to be analyzed, and extracting a feature sequence from the target sentence;
inputting the characteristic sequence into a pre-trained sentence emotion analysis model to obtain the emotion type of the target sentence at the target view angle;
and combining the emotion types of all the sentences determining the emotion types at the target view angle to obtain the emotion types of the text to be analyzed at the target view angle.
Wherein the extracting the feature sequence in the target sentence comprises:
performing word segmentation processing on the target sentence to obtain a plurality of words;
constructing a feature sequence based on the multiple word segments, wherein the feature sequence comprises: each participle, part of speech corresponding to each participle, identification corresponding to each participle, a predicate of a target perspective word, and a position of each participle relative to the target perspective word or the predicate of the target perspective word;
the identification comprises: view identification, attribute identification, rating identification, or other identification.
Wherein, the performing statistical analysis on the text to be analyzed according to the target perspective words comprises:
if the title of the text to be analyzed does not contain the target visual angle words and the expression words of the target visual angle words, the number of sentences of the text to be analyzed is larger than a first preset threshold value, and the number of sentences containing the target visual angle words and/or the expression words of the target visual angle words in the text is smaller than a second preset threshold value, determining that the emotion type of the whole text to be analyzed at the target visual angle is neutral;
if the emotion type of the whole text to be analyzed in the target view angle is not neutral, counting the number of view angles contained in a first sentence containing the target view angle in the text to be analyzed; and if the number of the visual angles is larger than a third preset threshold value, determining that the emotion type of the first sentence at the target visual angle is neutral.
Wherein the determining of the target sentence in the sentences of which the emotion types at the target view angle are not determined in the text to be analyzed comprises:
if the text to be analyzed only includes the target perspective words and/or the expression words of the target perspective words and the number of the sentences including the expression words of the target perspective words and/or the expression words of the target perspective words is smaller than a fourth preset threshold value, if the sentences including the expression words of the target perspective words and/or the expression words of the target perspective words include the titles of the text to be analyzed and the number of the sentences of the text to be analyzed is smaller than a fifth preset threshold value, taking all the sentences which are not determined to be in the emotion types at the target perspective in the text to be analyzed as target sentences.
Wherein the determining of the target sentence in the sentences of which the emotion types at the target view angle are not determined in the text to be analyzed comprises:
if the text to be analyzed only includes the target perspective word and/or the expression word of the target perspective word, and the number of the sentences including the target perspective word and/or the expression word of the target perspective word is less than a fourth preset threshold, then, if the sentences including the expression word of the target perspective word and/or the target perspective word do not include the title of the text to be analyzed, and the number of the sentences of the text to be analyzed is less than a sixth preset threshold, the sentences which are not determined to be in the target perspective and have a length greater than a seventh preset threshold in the text to be analyzed are taken as the target sentences.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both permanent and non-permanent, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include transitory computer readable media (transmyedia) such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (7)
1. A text emotion analysis method is characterized by comprising the following steps:
identifying words used for representing a target visual angle to be analyzed in the text to be analyzed as target visual angle words;
performing statistical analysis on the text to be analyzed according to the target view angle words to judge whether the emotion types of the whole text to be analyzed at the target view angle are neutral, and/or judging whether the emotion types of sentences containing the target view angle words and/or expression words of the target view angle words in the text to be analyzed at the target view angle are neutral;
if at least one judgment result is negative, determining a target sentence in the sentence which is not determined to be in the emotion type of the target visual angle in the text to be analyzed, and extracting a characteristic sequence from the target sentence;
inputting the characteristic sequence into a pre-trained sentence emotion analysis model to obtain the emotion type of the target sentence at the target view angle;
merging the emotion types of all the sentences determining the emotion types at the target view angle to obtain the emotion types of the text to be analyzed at the target view angle;
the performing statistical analysis on the text to be analyzed according to the target perspective words comprises:
if the title of the text to be analyzed does not contain the target visual angle words and the expression words of the target visual angle words, the number of sentences of the text to be analyzed is greater than a first preset threshold value, and the number of sentences containing the target visual angle words and/or the expression words of the target visual angle words in the text is less than a second preset threshold value, determining that the emotion type of the whole text to be analyzed at the target visual angle is neutral;
if the emotion type of the whole text to be analyzed in the target view angle is not neutral, counting the number of view angles contained in a first sentence containing the target view angle in the text to be analyzed; if the number of the visual angles is larger than a third preset threshold value, determining that the emotion type of the first sentence at the target visual angle is neutral;
determining a target sentence in the sentences of which the emotion types at the target view angle are not determined in the text to be analyzed, wherein the determining of the target sentence comprises the following steps:
if the text to be analyzed only includes the target perspective words and/or the expression words of the target perspective words, and the number of the sentences including the expression words of the target perspective words and/or the target perspective words is smaller than a fourth preset threshold, if the sentences including the expression words of the target perspective words and/or the target perspective words include the titles of the text to be analyzed, and the number of the sentences of the text to be analyzed is smaller than a fifth preset threshold, all the sentences which are not determined to be in the target perspective in the text of the text to be analyzed are taken as target sentences.
2. The method of claim 1, wherein the extracting the feature sequence in the target sentence comprises:
performing word segmentation processing on the target sentence to obtain a plurality of words;
constructing a feature sequence based on the plurality of word segments, wherein the feature sequence comprises: each participle, part of speech corresponding to each participle, identification corresponding to each participle, a statement word of a target view word, and the position of each participle relative to the target view word or the statement word of the target view word;
the identification comprises: view identification, attribute identification, rating identification, or other identification.
3. The method of claim 1, wherein determining a target sentence among the sentences of which the emotion classification in the target view is not determined in the text to be analyzed comprises:
if the text to be analyzed only includes the target perspective word and/or the expression word of the target perspective word, and the number of the sentences including the target perspective word and/or the expression word of the target perspective word is less than a fourth preset threshold, then, if the sentences including the expression word of the target perspective word and/or the target perspective word do not include the title of the text to be analyzed, and the number of the sentences of the text to be analyzed is less than a sixth preset threshold, the sentences which are not determined to be in the target perspective and have a length greater than a seventh preset threshold in the text to be analyzed are taken as the target sentences.
4. A text emotion analysis device, comprising:
the identification module is used for identifying target visual angle words used for representing target visual angles to be analyzed in the text to be analyzed;
the analysis module is used for carrying out statistical analysis on the text to be analyzed according to the target view angle words so as to judge whether the emotion type of the whole text to be analyzed at the target view angle is neutral or not, and/or judge whether the emotion type of a sentence, which contains the target view angle words and/or expression words of the target view angle words, in the text to be analyzed at the target view angle is neutral or not;
the extraction module is used for determining a target sentence in the sentence which is not determined to be the emotion type of the target view angle in the text to be analyzed and extracting a feature sequence from the target sentence if at least one judgment result is negative;
the classification module is used for inputting the characteristic sequence into a pre-trained sentence emotion analysis model to obtain the emotion type of the target sentence at the target view angle;
the merging module is used for merging the emotion types of all the sentences determining the emotion types at the target view angle to obtain the emotion types of the text to be analyzed at the target view angle;
the analysis module is specifically configured to:
if the title of the text to be analyzed does not contain the target visual angle words and the expression words of the target visual angle words, the number of sentences of the text to be analyzed is greater than a first preset threshold value, and the number of sentences containing the target visual angle words and/or the expression words of the target visual angle words in the text is less than a second preset threshold value, determining that the emotion type of the whole text to be analyzed at the target visual angle is neutral;
if the emotion type of the whole text to be analyzed at the target visual angle is not neutral, counting the number of visual angles contained in a first sentence containing the target visual angle in the text to be analyzed; if the number of the visual angles is larger than a third preset threshold value, determining that the emotion type of the first sentence at the target visual angle is neutral;
the extraction module is used for, when the target sentence is not determined in the sentences of the emotion types of the target view angle in the text to be analyzed,:
if the text to be analyzed only includes the target perspective words and/or the expression words of the target perspective words, and the number of the sentences including the expression words of the target perspective words and/or the target perspective words is smaller than a fourth preset threshold, if the sentences including the expression words of the target perspective words and/or the target perspective words include the titles of the text to be analyzed, and the number of the sentences of the text to be analyzed is smaller than a fifth preset threshold, all the sentences which are not determined to be in the target perspective in the text of the text to be analyzed are taken as target sentences.
5. The apparatus according to claim 4, wherein the extraction module, when the feature sequence in the target sentence, is specifically configured to:
performing word segmentation processing on the target sentence to obtain a plurality of words;
constructing a feature sequence based on the multiple word segments, wherein the feature sequence comprises: each participle, part of speech corresponding to each participle, identification corresponding to each participle, a predicate of a target perspective word, and a position of each participle relative to the target perspective word or the predicate of the target perspective word;
the identification comprises: a view identifier, an attribute identifier, an evaluation identifier, or other identifier.
6. A storage medium, comprising a stored program, wherein when the program runs, a device on which the storage medium is located is controlled to execute the text sentiment analysis method according to any one of claims 1-3.
7. A processor, characterized in that the processor is configured to run a program, which when running executes the method for text sentiment analysis according to any one of claims 1-3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811159908.1A CN110969011B (en) | 2018-09-30 | 2018-09-30 | Text emotion analysis method and device, storage medium and processor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811159908.1A CN110969011B (en) | 2018-09-30 | 2018-09-30 | Text emotion analysis method and device, storage medium and processor |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110969011A CN110969011A (en) | 2020-04-07 |
CN110969011B true CN110969011B (en) | 2023-04-07 |
Family
ID=70028956
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811159908.1A Active CN110969011B (en) | 2018-09-30 | 2018-09-30 | Text emotion analysis method and device, storage medium and processor |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110969011B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105893582A (en) * | 2016-04-01 | 2016-08-24 | 深圳市未来媒体技术研究院 | Social network user emotion distinguishing method |
CN106528528A (en) * | 2016-10-18 | 2017-03-22 | 哈尔滨工业大学深圳研究生院 | A text emotion analysis method and device |
CN108470061A (en) * | 2018-03-26 | 2018-08-31 | 福州大学 | A kind of emotional semantic classification system for visual angle grade text |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8930187B2 (en) * | 2012-01-03 | 2015-01-06 | Nokia Corporation | Methods, apparatuses and computer program products for implementing automatic speech recognition and sentiment detection on a device |
-
2018
- 2018-09-30 CN CN201811159908.1A patent/CN110969011B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105893582A (en) * | 2016-04-01 | 2016-08-24 | 深圳市未来媒体技术研究院 | Social network user emotion distinguishing method |
CN106528528A (en) * | 2016-10-18 | 2017-03-22 | 哈尔滨工业大学深圳研究生院 | A text emotion analysis method and device |
CN108470061A (en) * | 2018-03-26 | 2018-08-31 | 福州大学 | A kind of emotional semantic classification system for visual angle grade text |
Also Published As
Publication number | Publication date |
---|---|
CN110969011A (en) | 2020-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106815192B (en) | Model training method and device and sentence emotion recognition method and device | |
CN107391545B (en) | Method for classifying users, input method and device | |
CN107704512A (en) | Financial product based on social data recommends method, electronic installation and medium | |
CN109460455A (en) | A kind of Method for text detection and device | |
US11144594B2 (en) | Search method, search apparatus and non-temporary computer-readable storage medium for text search | |
CN110569502A (en) | Method and device for identifying forbidden slogans, computer equipment and storage medium | |
CN109271489A (en) | A kind of Method for text detection and device | |
CN106372956B (en) | Method and system for identifying intention entity based on user search log | |
CN115392237B (en) | Emotion analysis model training method, device, equipment and storage medium | |
CN112926471A (en) | Method and device for identifying image content of business document | |
CN111291551B (en) | Text processing method and device, electronic equipment and computer readable storage medium | |
CN107291686B (en) | Method and system for identifying emotion identification | |
CN113297482B (en) | User portrayal describing method and system of search engine data based on multiple models | |
CN110969011B (en) | Text emotion analysis method and device, storage medium and processor | |
CN111222051A (en) | Training method and device of trend prediction model | |
CN115617998A (en) | Text classification method and device based on intelligent marketing scene | |
CN109359274A (en) | The method, device and equipment that the character string of a kind of pair of Mass production is identified | |
CN117271713A (en) | Associated object recognition method, associated object recognition device, electronic equipment and storage medium | |
Karim et al. | Classification of Google Play Store Application Reviews Using Machine Learning | |
CN108021548A (en) | A kind of recognition methods of affective characteristics and device | |
CN109558582B (en) | Visual angle-based sentence emotion analysis method and device | |
CN111353688A (en) | User resource allocation method and device | |
CN112579768A (en) | Emotion classification model training method, text emotion classification method and text emotion classification device | |
Sumathi et al. | Sentiment Analysis on Feedback Data of E-commerce Products Based on NLP | |
CN111400577A (en) | Search recall method and device |
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 |