40 results about "Comment analysis" patented technology

The invention discloses a comment analyzing method and system based on a term vector and relates to the technical field of emotion analysis, natural language processing and the like. A machine is utilized to analyze the comment, automatic user comment analysis is made by using the machine, and the working efficiency is improved. The method is characterized in that user comments are collected to form a comment corpus, each comment in the comment corpus is converted into the sentence vectors with identical dimension, a plurality of comment types are set, each comment is labeled with the corresponding type according to the labels which are input manually, a classifier is trained with the sentence vectors as the input and the comment type that each sentence vector corresponds to as the output, a new comment is acquired and converted into the sentence vector, and the sentence vector that the new comment corresponds to is input into the classifier to obtain the comment type of the new comment.

The invention provides a comment analysis method based on word vectors and syntactic characteristics in the field of data analysis. The comment analysis method comprises the steps of obtaining commodity page comment data of an e-commerce website; preprocessing the acquired target data set; extracting a appendix lexical set provided by Hownet and NTU to form a basic emotion dictionary; carrying outword vector training on the obtained preprocessed data set through a Word2Vec tool; establishing a probability transfer matrix by using the semantic similarity matrix; carrying out core sentence rule-based processing on the obtained commodity comment text; carrying out preprocessing on the obtained text without the redundancy; performing part-of-speech extraction (commodity attributes, negative words, degree words and sentiment words) evaluation matching on the obtained dependency relationship pairs; combining the evaluation matching pair with an emotion dictionary, subjecting evaluation objects to appraisal value calculation and quality sorting, and finally, realizing the evaluation objects through a visual interaction interface, so that accurate, real-time, automatic and convenient processing and analysis on commodity comment data are realized, and the method can be used in an e-commerce platform.

A method includes determining a plurality of topics corresponding to descriptive text and to comments concerning the descriptive text, wherein each of a number of sets of the plurality of topics comprise a similar topic and one or more supplemental topics. The method includes determining by a computer system probabilities for words, where the probabilities are that the words belong to individual ones of the topics. The method further includes generating, by the computer system and based on comments and the probabilities, a content summary comprising a plurality of comment snippets having positive and negative sentiments toward corresponding ones of the similar or supplemental topics in the sets of topics. Each topic has corresponding comment snippets having positive and negative sentiments. The method includes outputting by the computer system at least a portion of the plurality of topics and corresponding comment snippets. Apparatus and computer program products are also disclosed.

The invention discloses a fine-grained emotion analysis improvement method based on emotion word embedding. The method comprises: obtaining of a text data set, manual labeling and preprocessing; calculating a semantic word vector corresponding to each word in the text data set; obtaining a set of sentiment words; calculating a group of emotion phrases corresponding to each word in the training data set; and calculating an emotion word vector corresponding to each word in the training data set; constructing emotion word embedding corresponding to each word in the training data set; and trainingthe classifier to obtain a fine-grained emotion analysis model. According to the method, priori sentiment knowledge is combined with a word embedding model, and sentiment word embedding suitable forfine-grained sentiment analysis is constructed for sentiment analysis. The emotion information of the word level can be better identified, the emotion of the user can be described more accurately in afine-grained mode, and the method can be used for fine-grained emotion analysis tasks such as consumption habit analysis of the user and comment analysis of the user on commodities.

The invention provides a label system accurate recommendation method based on user comment analysis. The interest model is constructed according to the user-commodity-label ternary relation, the accurate recommendation method more suitable for the label system is obtained, and in view of the problems that label information data of users in the label system generally has data sparseness and the user similarity calculated by using sparse data is low in accuracy, user comment data is creatively introduced, text analysis on user comment information is carried out, Chinese word segmentation and keyword extraction on the comment information are carried out, the extracted keywords are taken as pseudo tags, user labels are extracted, the label information data is expanded, the problem of label information data sparseness is solved, meanwhile, based on the fact that user comment information contains user preferences, value assignment calculation is conducted on emotion words in the comment information, the score value of a user for a commodity is obtained from user comments, the obtained score value information is used for further improving a label algorithm, and the accuracy of a recommendation result is improved.

The invention discloses a restaurant user comment analysis system and method based on emotion analysis, and the system comprises a user comment obtaining module which is used for carrying out data crawling on different platforms; a data preprocessing module which is used for preprocessing the data; a feature extraction module which is used for performing feature extraction on the preprocessed datato obtain a word feature vector and an aspect level vector; a pre-training language model module which is used for training a language model; an algorithm processing module which is used for inputting the word feature vectors and the aspect level vectors subjected to feature extraction into an algorithm model for calculation and analysis; a result display module which is used for displaying the calculation analysis result and the system operation condition; and an authority authentication module which enters different interfaces through different identity authentications. According to the method, the built-in user comment text data and the data from the software platform can be processed, a simpler and easier-to-understand result presentation interface is provided, and the understanding difficulty of users and merchants can be effectively reduced.

The invention discloses a dynamic internet comment analysis method. The dynamic internet comment analysis method comprises the following steps: S1, collecting opinion comment texts of all commentators, and transmitting opinions of the commentators to a dynamic analysis platform; S2, uploading information data of a seller, in response to the opinion comment texts of the commentators, to the dynamic analysis platform; S3, associating information with the dynamic analysis platform by a user; S4, integrating the information in the steps S1, S2 and S3 by the dynamic analysis platform, and performing analysis and excavation. Through the mode, the dynamic internet comment analysis method provided by the invention has the advantages that the source of comment opinions is authoritative, professional, objective, fair, correct and wide, and comprehensively covers experts, users, detectors, industrial elites and the like, misdirection and overspreading of various conventional comment websites and APPs are effectively avoided, and multi-class authoritative and objective popular opinions can be obtained.

The invention belongs to the technical field of natural language processing, and discloses a fine-grained attribute analysis method in an e-commerce comment analysis scene, which comprises the following steps of: respectively extracting target attributes and emotions of e-commerce comments by using two isomorphic and independently trained sequence labeling models; using a relationship classification model to perform relationship classification on the target attributes and emotion extracted by the model in the step 1; wherein the training data of the model is carried out by using an efficient labeling strategy, and the data labeled by using the strategy is used for training the models in the step 1 and the step 2; the rule matching relation module is used for integrating target attributes and emotions output by all the models; and the post-processing module is used for integrating the output of the rule matching relationship module and the output of the relationship classification model. Through the solution provided by the invention, on the premise of ensuring the accuracy of the fine-grained emotion prediction result, the complexity of the fine-grained emotion annotation task is simplified, and the annotation efficiency is improved.

The invention discloses a competitive product analysis method and system based on big data, a storage medium and electronic equipment. The method belongs to the technical field of big data and competitive product analysis. The technical problem to be solved by the invention is how to analyze and mine emotional trends of netizens on specified products and competitive products thereof by using big data to realize competitive product analysis. The adopted technical scheme is as follows: ;according to the method, relying on completely public social network platform comment data, the emotion trendsof netizens to specified products and related competitive products are mined from network comments by analyzing massive comments, and the attitude opinions of the netizens to the products are obtained by performing multiple analysis modes including emotion analysis on the comments, so that the purpose of competitive product analysis is achieved. The system comprises a product and related competitive product determination unit, a topic generation and comment acquisition and storage unit, a comment denoising and cleaning unit, a comment word segmentation and stop word removal unit, a comment emotion score acquisition and tagging unit and a comment analysis and product optimization and upgrading unit.

The invention discloses a false comment method based on a convolutional neural network, and the method combines a mainstream short text feature mining model, and enables the convolutional neural network to be used for the recognition of false commodity evaluation in the field of short text processing. The method comprises the following steps: firstly, selecting a sample data set, taking 80% of sample data as a training set, and taking 20% of samples as a test set; and inputting the training set into a Word2Vec model to obtain a word vector dimension; and then, through convolution calculation and feature extraction, enabling the output result of the model to have relatively high accuracy in precision, recall rate and f1-score. This indicates the feasibility of applying the convolutional neural network model to false comment recognition.

The invention provides a live streaming comment analysis method and device, electronic equipment and a storage medium. The method comprises the steps of: obtaining a bullet screen text and a comment text generated in a video live streaming process, so as to obtain comment data; performing word segmentation on the comment data to obtain a plurality of words; performing clustering analysis and category labeling on the words to obtain a plurality of category labels; and screening commodity names from the plurality of category labels to obtain a plurality of commodity names. In the implementationprocess, the comment data is subjected to word segmentation, clustering analysis and category labeling, so that the plurality of category labels can be obtained, and the commodity names in the commentdata are screened out from the category labels, so that the problem of low efficiency of obtaining commodity names expected to be subjected to next live streaming or later live streaming by a live streaming user is effectively improved.

The invention relates to a comment analysis method and device. The method comprises the following steps: extracting new words from comments aiming at a specified object; performing word segmentation processing on the comments according to the new words to obtain word segmentation results of the comments; performing syntactic analysis on the word segmentation results of the comments to obtain syntactic analysis results of the comments; according to the condition that the syntactic analysis result of the comments contains seed words or new words, determining viewpoint types corresponding to thecomments; and according to the syntactic analysis result of the comments corresponding to the viewpoint categories, determining the emotional tendency of the specified object in each viewpoint category. According to the method, fine-grained sentiment analysis can be carried out on the specified object, the sentiment tendency of the specified object in each viewpoint category can be accurately determined, a business worker can be helped to know the comment angle and appraisal attitude of the vast users on the specified object, and the value of the comment information is fully mined.

The invention relates to a song comment analysis method and system. The method comprises the steps of selecting samples of song comments, and performing corresponding marking on the samples to obtainmarked samples; according to predetermined rules, dividing the marked samples into training set samples and test set samples; performing word segmentation processing on the training set samples and the test set samples, and training a processed text through a training model to obtain corresponding text vectors; and training the text vectors in the training set samples through a training classification model to obtain an optimized training classification model, and analyzing the song comments in the test set samples through the optimized training classification model. The optimized training classification model not only solves the problem of context information missing of a song comment analysis method at present and the problem of excessively large data set scale, but also greatly improvesthe accuracy of song comment analysis.

The invention provides a catering comment analysis method, and the method comprises the following steps: carrying out the word dividing of the comment information of a restaurant; carrying out the recognition of obscene words of a file after word dividing; deleting the comment if the obscene words are recognized. The method can achieve the analysis and processing of comments.

The invention relates to a comment analysis method and device. The comment analysis method comprises the following steps: extracting a first group of new words from comments for a specified object; according to the first group of new words, performing word segmentation processing on each comment aiming at the appointed object to obtain a word segmentation result aiming at each comment aiming at the appointed object; inputting a word segmentation result of the comment into a prediction model for any comment of the specified object to obtain a viewpoint category corresponding to the comment anda viewpoint part in the word segmentation result of the comment; and according to the viewpoint part in the word segmentation result of the comment corresponding to each viewpoint category, determining the emotional tendency of the specified object in each viewpoint category. According to the comment analysis method, fine-grained emotion analysis can be carried out on the specified object; the emotional tendency of the specified object in each viewpoint category can be accurately determined; a business worker can be helped to know the comment angle and appraisal attitude of the vast users on the specified object; and the value of the comment information is fully mined.

The invention discloses an inspection agency dynamic comment analysis method comprising the following steps: S1, inspection agency information / opinion is collected and transmitted to a dynamic analysis platform; S2, information data of a seller responds to the inspection agency information / opinion, and the information data is uploaded to the dynamic analysis platform; S3, a user associates information with the dynamic analysis platform; and S4, the dynamic analysis platform synthesizes the information in S1, S2 and S3, carries out analysis and mining, expresses attribute features in words, and finally, summarizes the emotional tendencies of the corresponding attribute features in a classified way. According to the inspection agency dynamic comment analysis method provided by the invention, the source of inspection agency opinion is authoritative, professional, objective, fair, correct and comprehensive, and the phenomenon that the inspection result is inaccurate is avoided effectively.

The invention relates to a multi-platform collaborative new media content monitoring management system based on big data. The multi-platform collaborative new media content monitoring management system comprises a user terminal and a detection cloud platform. The detection cloud platform comprises a comment analysis module, an emotion analysis module, a relation construction module, a content verification module and a news rating module. The detection cloud platform obtains first comment information and second comment information of each social user according to the comment content published by each social user, the relation construction module obtains the comment relevancy between the social users on the basis of the first comment information and the second comment information and constructs corresponding association topological graphs for the different social users; the content verification module is used for recognizing focus content of the news content on the basis of the association topological graphs and the deep semantic features of the social users and verifying whether the focus content is false news content or not; and the news rating module is used for rating the influence of the false news content based on the user influence of the media user and the forwarding data and the browsing data of the false news content.

The invention discloses a movie comment emotion analysis method and device, belongs to the field of data analysis, and particularly relates to a comment analysis method and device. The objective of the invention is to solve the problem of low accuracy in sentiment analysis of movie comments by using an existing sentiment analysis method. The method comprises the following steps: firstly, carryingout pretreatment such as word segmentation and syntactic analysis on film review corpora; Judging whether the words belong to words in a film review attribute feature word dictionary or not, inquiringand judging brother nodes of father nodes of the words, and obtaining the words corresponding to the nodes closest to the words in the sentences as emotion words; Traversing the W-SVM sentiment dictionary and the SO-VEC sentiment dictionary, and judging which category in the sentiment dictionary the emotional word belongs to, and recording the emotional value of the word; And finally updating theword sentiment value according to the negative word number between the word and the sentiment word; And finally, determining the movie review fine-grained emotion according to the word emotion values. The method is suitable for emotion analysis of movie comments.

The embodiment of the invention discloses a product comment information processing method and system combined with big data, and the method can achieve the precise positioning and obtaining of the optimization suggestion mining information of an active user comment text set under a plurality of types of product feedback suggestion themes during the recognition and mining of optimization suggestions. According to the method, optimization suggestion recognition omission of individual product feedback suggestion themes of the active user comment text set can be reduced to a certain extent, so that the precision and integrity of optimization suggestion recognition of the active user comment text set can be improved as much as possible, and in addition, due to the fact that the core process of optimization suggestion recognition is carried out on a big data processing system side, the optimization suggestion recognition efficiency is improved. Therefore, the comment analysis resource overhead of a digital product platform system, especially a digital product platform system with relatively weak computing power, can be reduced.

The invention discloses a user comment analysis method based on text and picture contents. The method comprises the following steps: 1, inputting weight data of current users participating in ranking; 2, establishing a user ranking method according to the weight data; 3, calculating the user ranking score; 4 storing the user data in a dictionary and storing the ranking score in a skip list, (5) calculating the user ranking according to the skip list and obtaining the corresponding user data, and (6) displaying the user ranking in a game. The ranking system can quickly rank mass users of a farm game. According to the system, the ranking of the users in the game takes the money input ratio and the consumption time input ratio of the consumer products of the users as the ranking basis, and meanwhile, the skip list and the dictionary are taken as the bottom layer data structure of the ranking system, so that the problems of too low fairness of user ranking under multiple weights and low ranking efficiency under mass data can be effectively solved.

The present disclosure discloses a product recommendation method and system based on multimodal product review analysis, including: obtaining review information of a product; performing data preprocessing on the obtained review information of a product; judging whether there is an image in the review information, and if there is an image , then extract the emotional label of the image from the image; judge whether there is a video in the comment information, if there is a video, extract the audio from the video, and convert the audio into text; judge whether there is audio in the comment information, if there is audio, then Convert the audio to text; judge whether there is text in the comment information, and if there is text, integrate the text in the comment information with the converted text to obtain the integrated text; extract the emotional label of the text from the integrated text ; Input the emotional label of the image and the emotional label of the text into the pre-trained neural network, and output the recommended label of the current product.

The invention discloses a comment analysis method and device based on semantic perception. The comment analysis method comprises the steps of collecting a comment text of a target application; dividing each evaluation text into at least one sentence; comment attributes of the comment text are extracted; splicing the vector of each word in the sentence with the vector of the comment attribute; based on a splicing result, calculating a probability vector of a word BIO tag to obtain defect characteristics in the evaluation text; and clustering the defect characteristics to obtain a comment analysis result. According to the method, a defect characteristic extraction task is modeled as named entity recognition, and the accuracy of phrase recognition is improved by introducing a defect characteristic recognition model of comment attributes.

The invention discloses an internet user comment analysis method based on big data text analysis. The method comprises the following steps: obtaining and sending user evaluation information; performing data cleaning processing on the user evaluation information to obtain and send target pure text information; extracting user emotion data in the target pure text information, classifying the user emotion data according to a set rule, and generating and sending a classification report; performing word segmentation operation on the text information in the classification report by using a Chinese word segmentation framework to obtain and send word segmentation text data; performing word vectorization operation on the word segmentation text data, and removing interference information to obtain vectorized words; and performing operation on the vectorized words by utilizing a naive Bayes algorithm, and outputting and displaying an operation result. The invention further discloses an internet user comment analysis system based on big data text analysis. According to the invention, the user comment analysis efficiency and accuracy can be effectively improved.

The invention relates to the technical field of comment analysis, and provides a product feature clustering method based on adjacent word similarity, which comprises the following steps: S1, extracting product features based on product comments, and putting the product features into a product feature set; s2, calculating the similarity of two product features in the product feature set to form a similarity matrix; and S3, clustering the product features by using a hierarchical clustering method to form a tree-shaped product feature clustering structure. A product feature similarity measurementmethod of adjacent word similarity is adopted, and clustering is performed based on the layering of product features so as to better summarize customer comments.

The invention discloses a merchant comment analysis system based on machine learning. The merchant comment analysis system comprises a data processing layer, a machine learning algorithm layer and a user service layer, wherein the data processing layer is used for acquiring platform data, performing preprocessing, text word segmentation and word cloud generation on the platform data, and outputting an evaluation data set; the machine learning algorithm layer is used for performing model training, sentiment analysis and clustering analysis by using a comment data set, and outputting the positive sentiment degree and merchant classification labels of the comment data; and the user service layer provides merchant recommendation, search, classification, merchant data analysis and comment data analysis information for users. According to the invention, a machine learning algorithm is adopted to analyze the comment data of users, quantify the sentiment data of the users and mine the emotion and attitude of the users in the comments, so merchant evaluation can be refined, the shortages of platform scoring are made up, a multi-dimensional scoring model is constructed, and a more accurate merchant evaluation system is realized.

The invention discloses a Chinese commentary analysis method applicable to collecting a Chinese 'pseudo comment' corpus by analyzing Chinese comments of users to determine whether the comments can be used as corpus or not. The method includes the steps that a user submits a comment to a website, the front-end of the website sends an analysis request to a control center, the control center transfers the comment to an analysis unit, the analysis unit performs theme classification analysis on the comment, a word classification server performs word classification and part-of-speech tagging, the analysis unit sequentially performs syntactic analysis and sentiment analysis, and a data center saves analysis conclusions to a user comment table. According to the Chinese commentary analysis method, the control center can directly exclude unqualified corpus through thematic analysis and the analysis unit sequentially performs syntactic analysis and sentiment analysis on the comment of the user so as to effectively draw conclusions of emotional tendencies of the Chinese comment and improve the accuracy of an analysis system, and then an administrator can only view the comment in positive tendencies to determine whether the comment meets the requirements or not.

The invention discloses a video comment viewpoint processing method and device, and the method comprises the steps: extracting multi-party viewpoints of a video comment through the viewpoint mining and tendency understanding of the video comment, automatically generating a voting strategy, and then displaying the viewpoint tendency of a user in real time through the analysis of the video comment and voting data. According to the processing mode of generating the user voting scheme based on the video comments, comment analysis is provided for the user, meanwhile, attention of the user is attracted, the user can know the opinion tendency in the comments and the opinion proportion of multiple parties, more users are guided to join in interaction, and the user experience of the platform is improved.

The invention provides a mobile application comment analysis and mining method based on metroeconomics, relates to fusion and analysis and mining of data generated by different types of users, and particularly relates to the establishment of a metroeconomics model based on mass comments uploaded by the users in a mobile application market so as to generate user feedback with guiding significance.The method not only can deeply analyze the actual needs of the user through feature extraction, but also can describe the influences of different factors on the downloading amount of the mobile application by utilizing the metrological model, and can meet different user needs at the same time.