273 results about "Opinion analysis" patented technology

The invention provides a smart phone customer service system based on speech analysis. Firstly, real-time recording and effectiveness detection are performed on conversation speech caused when a client calls human customer service; then related individual information of the client is extracted, voiceprint recognition is performed on client speech in the conversation speech, and the verification is performed and the conversation speech is used as client identity at the time of consulting complaints to be recorded; meanwhile, speech content recognition is performed on the conversation speech to save as written record, and text public opinion analysis is performed on the written record; speech emotion analysis is performed on the conversation speech to record speech emotion data; the effect of the consulting complaints this time is analyzed by combination of the result of the text public opinion analysis and the emotion data; final grade of the customer service is obtained by combination of the analysis results and traditional grading evaluation to perform feed-back assessment. Compared with traditional customer service systems, the smart customer service system is capable of effectively increasing service qualities of the client and achieving objectified management of customer service evaluation.

System and method for analysis of an opinion expressed in documents on a particular topic computes opinion strength on a continuous numeric scale, or qualitatively. A variety of opinion scoring techniques are plugged in to score opinion expressing words and sentences in documents. These scores are aggregated to measure the opinion intensity of documents. Multilingual opinion analysis is supported by capability to concurrently identify and visualize the opinion intensity expressed in documents in multiple languages. A multi-dimensional representation of the measured opinion intensity is generated which is agreeable with multi-lingual domain.

A regional public opinion monitoring and decision-making auxiliary system based on big data comprises an information acquisition and storage module, a data pre-processing module, a big data public opinion analysis module, a public opinion monitoring, early warning and decision-making auxiliary module and a background management module, wherein the information acquisition and storage module is used for carrying out structuralized storage management on acquired public opinion source information to form a regional big data public opinion knowledge base which is updated in real time; the data pre-processing module is used for pre-processing data in the regional big data public opinion knowledge base to form a complete and ordered data set; the big data public opinion analysis module is used for carrying out public opinion analysis and tendency prediction on hot topics with appointed conditions and the like to obtain a public opinion analysis and tendency prediction result; the public opinion monitoring, early warning and decision-making auxiliary module is used for carrying out real-time monitoring and tracking, management and persuasion on customized public opinions dug and analyzed by sensitive word pairs in a pre-defined sensitive word bank, and announcing the customized public opinions to a decision maker in manners of in-station messages, short messages and mails; and the background management module is used for carrying out public opinion information classification management, user and authority management, keyword management, acquisition management, content management, special topic management and analysis report management.

The invention discloses an internet public opinion analysis method. The method includes the following steps: determining a public opinion event, and collecting public opinion articles related to the public opinion event; preprocessing the collected public opinion articles, and obtaining a vocabulary collection in the public opinion articles to characterize the public opinion articles; performing clustering analysis on the vocabulary collection by adopting a clustering algorithm, generating a plurality of viewpoints of the public opinion event, and calculating a word vector of the viewpoints; extracting a core topic from the vocabulary collection contained in the viewpoints; calculating an emotional score of the viewpoints through an emotional scoring model, and calculating the level of popularity of the viewpoints; and calculating a public opinion index of the viewpoints according to the emotional score and the level of popularity, and determining that the viewpoint of which the absolute value of the public opinion index is greater than a preset threshold is an abnormal viewpoint, generating early warning information according to the abnormal viewpoint and the core topic, and outputting the early warning information. The invention also provides an internet public opinion analysis device and a computer readable storage medium. According to the scheme of the invention, the monitoring and early warning capabilities for public opinions can be improved.

The invention discloses a multiple features fused bidirectional recurrent neural network fine granularity opinion mining method. The method comprises the following steps of: capturing comment data of a specific website through internet and carrying out labelling and preprocessing on the comment data to obtain a training sample set; carrying out training by using a Word2Vec or Glove model algorithm to obtain word vectors of the comment data; carrying out vectorization after carrying out part of speech labeling, dependence relationship labeling and the like; and inputting the vectors into a bidirectional concurrent neural network to construct a bidirectional recurrent neural network fine granularity opinion mining model. According to the method, attribute words in fine granularity opinion mining is extracted and emotional polarity judgement is carried out through the training of a model, so that plenty of model training time is further saved and the training efficiency is improved; no professionals are required to carry out manual extraction on the attribute words, so that a lot of manpower cost is saved; and moreover, the model can be trained by using a plurality of data sources, so that cross-field fine granularity opinion analysis can be completed, thereby solving the problem of long-distance emotional element dependency.

The invention discloses an internet public opinion analysis method. The internet public opinion analysis method comprises the steps of firstly for selected and acquired events, partitioning a source text of a microblog and removing partition items unrelated to sentiments; secondly making statistics by adopting a statistic analysis tool to obtain an input of a sentiment classification model; and finally for the input, modeling related words, expressions and symbols capable of expressing the sentiments in the microblog content by using a classification algorithm, giving out comprehensive sentiment index assessment, obtaining sentiment categories, and performing public opinion monitoring and sentiment trend analysis. According to the method, the words, the expressions, the symbols and the like in the microblog are subjected to sentiment modeling, and the response situations of hot events in the microblog can be automatically classified and effectively monitored through sentiment index calculation, so that the public opinion risk can be effectively assessed and intemperate events can be prevented and controlled.

The invention discloses a text emotion analysis system based on deep learning. The system comprises an information collection module, an information pre-processing module, an emotion analysis module and an information display module, wherein the information collection module is used to collect comment information in each Internet resource website; the information pre-processing module is used to conduct classification, word segmentation, part-of-speech tagging, emotion information tagging processing and storage of the collected comment information; the emotion analysis module transforms the processed comment information into a phrase vector by a word representation model, a sentence module and a section and chapter model, and also inputs the phrase vector into the emotion classification model for emotion analysis; and the information display module is used to present emotion analysis results in a visualized manner. The system has the advantages that emotion orientation analysis can be conducted on the comment information; the analysis results can be presented to users in a visualized manner; and further public opinion analysis results or early warning can be provided to related departments such as enterprises or governments.

The invention discloses a time window based LDA microblog topic trend detection method and apparatus. The method comprises: extracting a topic word from a word set by utilizing an LDA model in each time window, and obtaining global topics; performing similarity computing on the global topics, and performing K-means clustering to obtain hot topics conforming to public opinion analysis; extracting feature words of each hot topic in each time window in sequence in combination with the hot topic through the LDA topic model; and in combination with results of the feature words, computing a popular value of the hot topic in each time window, and drawing a trend graph of the hot topic. The apparatus comprises a first acquisition module, a second acquisition module, an extraction module and a drawing module. According to the detection method and apparatus, the precision of microblog topic detection is improved, so that a trend index is more expressive, and a more accurate basis is provided for analyzing a hot topic trend.

The invention discloses a self-learning sorting method relating to multilingual data treatment, comprising the steps of extracting candidate emotion words by a first seed word Chinese or foreign language 'very', filtering stop words, and automatically obtaining a stop word list from a language database; simultaneously carrying out support or opposing clustering on the emotion words and emotion texts by a second seed word 'good' and a third seed word 'bad' or foreign languages 'good' and 'bad'; building an emotion classifier by semi-supervised learning, training the initial classifier by selecting convinced samples from a clustering result, and selecting new samples to be added into a training set by fusing emotion scores of the texts and the posterior probability of the classifier. According to the sorting method, the method facing multilingual opinion analysis is irrelevant with languages, a machine translation system and a large-scale bilingual dictionary are not needed, the emotion classifier is directly learned on a target language, the resource dependence is the smallest, and for each target language, only three seed words are needed and other priori knowledge is not needed.

The invention discloses an international capacity cooperation risk assessment and decision service system based on big data. The system comprises a distributed structure / unstructured database, a dataaccess layer which receives data collected by a front end through a standard communication interface and automatically inputs the data into the distributed structure / unstructured database according toa standard format, an information collection and pre-processing layer which collects data, pre-processes the collected data, mines and refines effective information, extracts information keywords andconstructs an index, an analyzing layer which identifies valuable information content from real-time information flow, clusters information for country risk, industry analysis and international public opinion analysis, and displays the information in the form of an information table, and an application layer which provides query and retrieval functions for query users, carries out multi-dimensional visualized output on the queried and retrieved information, and generates a corresponding report. According to the invention, information technological means are used to strengthen the risk prevention and control capability of Chinese enterprises to participate in the international capacity cooperation process.

The invention discloses a method for internet public opinion analysis, which particularly includes the following steps: matching key words and key sentence patterns in a text through an AC automat and a regular expression; expressing the text by a plurality of themes as per the matching results; and quickly and accurately analyzing and judging that whether a page belongs to public sentiment through calculating total weight of the page after the weight of each theme is set. According to the method provided by the invention, theme matching is adopted to substitute term matching in the simple internet public opinion analysis, complicated procedures such as clustering and classification are eliminated at the same time, so that various indexes can be integrated while a certain accuracy is guaranteed, high-speed internet public opinion analysis on various pages can be carried out, and the highest analysis speed is reached.

The invention relates to a network public opinion analysis method and system. The network public opinion analysis method comprises the steps: setting public opinion keywords of public opinion focus points, capturing and recording a data set which is related to the public opinion keywords and contains a plurality of pieces of network public opinion information according to the public opinion keywords, and carrying out the sentiment tendency analysis on the data set, wherein the step of carrying out sentiment analysis on the data set further comprises the following steps: a, preprocessing each frame of network public opinion information based on a pre-trained deep neural network language model to output a word vector matrix and / or a character vector matrix; b, inputting the word vector matrix and / or the character vector matrix into a word sequence fusion neural network, and performing public opinion feature extraction on each frame of network public opinion information; and repeating thestep a and the step b to establish a public opinion feature set of the data set, and outputting an emotion tendency analysis result of the public opinion on the attention point according to the public opinion feature set. The network public opinion analysis method and system can automatically monitor the internet public opinion in the specific professional field in real time.

The invention discloses a method for identifying microblog key users based on an improved Page Rank. The method comprises the steps that microblog information data are input, wherein the microblog information data comprise n microblogs; word segmentation is conducted on texts of the n microblogs; according to a word segmentation result, a reverse index structure is established, so that retrieval is conveniently conducted according to appointed keywords; according to the retrieved relevant microblog, forwarding hierarchy information of the microblog is extracted and a weighting directed graph is established, wherein the weighting directed graph is a forwarding network G; the forwarding network G is divided into a plurality of maximum connected subgraphs Gi; the Page rank algorithm is applied to each sub network Gi according to the parallelization computing technology; computing results of the sub networks are combined, so that ranking results of the whole network G are generated; the first m ranking results of the ranking results are selected, serve as the key users and are output. According to the method for identifying the microblog key users based on the improved Page Rank, the parallelization computing technology is adopted, a dynamic forwarding network of a microblog platform is ranked and computed in a big data environment, so that the key users in the information transmission process are identified, and the method is applied to the fields of network public opinion analysis and the like.

The invention discloses a method and system for radio and television public opinion analysis based on a deep learning technology. The method comprises the process of extraction and identification of public opinion information data based on the deep learning technology: first a data model is generated by repeated learning and training of plenty of material data (including images, videos, voices and optical character recognition (OCR) optical characters), and an image feature library is generated in combination with a computer vision technology (local feature grabbing and matching). To-be-monitored data (including audios, videos and pictures) is obtained from communication media such as cable digital television, wired radio television, satellite television and wireless radio television to serve as input of a deep learning data model, and the established data model is utilized to identify the to-be-monitored data. The plenty of material data forms a characteristic fingerprint database in combination with a computer vision technology, characteristic values of the to-be-monitored data are compared with data in the characteristic fingerprint database, and if the comparison is successful, it is thought that the to-be-monitored data contains public opinion information.

The invention discloses a microblog topic emotion analysis method based on topic influence. The method comprises the following steps: collecting personal information data and microblog text data of amicroblog user; Preprocessing the data; Using an LDA model to carry out theme clustering; Selecting microblog contents of the topics for emotion polarity labeling and performing training to generate an SVM classifier; Using an SVM classifier to obtain the emotional tendency of the unlabeled text; Calculating microblog user influence factor values and similarity between microblogs and topics, calculating emotion values of the whole topics by combining microblog emotion polarities and topic influence factors, and sequencing the topics according to the emotion values, so as to provide reference for public opinion analysis. The public opinion event and hot topic possibly emerging in the microblog text topics can be found in time, the topic public opinion degrees are ranked according to topic influence, the method can be applied to network public opinion supervision and emergent public opinion event monitoring, and the problems that an existing public opinion system is not supervised in time and blindness are solved.

The invention provides an analysis method of network public opinion, which comprises the following steps: determining preset keywords according to public opinion events, and collecting public opinionarticles related to the preset keywords; segmentation of the public opinion articles to obtain a vocabulary set corresponding to each public opinion article; a clustering algorithm being used for clustering analysis of the public opinion articles, a plurality of viewpoint topics of the public opinion events being generated, and emotion categories of the public opinion articles being determined according to the viewpoint topics; filtering out target public opinion articles about the public opinion event, and calculating public opinion scores of the target public opinion articles by using presetcalculation rules; determining a public opinion level corresponding to the target public opinion article; and, the warning information being generated and output according to the target public opinion article with the highest public opinion level and the corresponding viewpoint theme. The invention also provides an electronic device and a storage medium. The invention realizes the overall monitoring and early warning of the public opinion event, and is convenient for the user to make corresponding decision according to the public opinion event.

The invention discloses a method for detecting a 'sockpuppet' based on a 'similar view' network and an article author judging technique. The method comprises the three steps that first, the 'similar view' network is established according to the mutual information of virtual social users; then the 'similar view' network is cut according to the writing styles of the users; at last, a community discovery algorithm is used for conducting community dividing on the cut 'sockpuppet' network, and the IDs in the same community are considered as the 'sockpuppet' of a certain person. The method for detecting the 'sockpuppet' has the following advantages of following the actual significance of the 'sockpuppet' community, being capable of being applied to detecting the 'sockpuppet' under a real-time network environment, and increasing the effectiveness of community discovery. The method for detecting the 'sockpuppet' is mainly applied to the various fields of public opinion analysis in virtual space, 'sockpuppet' detection and the like.

The invention discloses a public transportation passenger satisfaction evaluation method and system based on a public opinion knowledge graph, and comprehensively considers a passenger satisfaction analysis method based on questionnaire survey and public transportation network public opinion. A network public opinion analysis technology, a natural language analysis method, a public transport knowledge graph, machine learning and other methods are applied to make a powerful supplement to traditional public transport passenger satisfaction analysis. The method mainly comprises the following steps: implementing text acquisition and processing analysis; completing modeling and storage of the public transport public opinion knowledge graph; acquiring and analyzing satisfaction information of'diving 'passengers; achieving public opinion index mapping, service score research and judgment and construction of a historical event method library. According to the method, text topic classificationand satisfaction index positioning are achieved, and corresponding relations between satisfaction and departments and responsibilities thereof are defined, so that powerful supplements can be providedfor traditional public transport passenger satisfaction analysis, and more scientific reference opinions can be provided for understanding public transport passenger demands and related department service shortages.

The invention relates to the field of data acquisition. Specifically, a knowledge graph is constructed by using a data crawling mode. The invention further discloses a crawler-based knowledge graph processing method and device, computer equipment and a storage medium. The method comprises the following steps of crawling data information in webpage information by utilizing a crawler system; performing data cleaning on the crawled data information; performing information completion on the crawled data information subjected to data cleaning; target information, and performing feature extraction on the target information, constructing a relationship classification model according to the extracted feature information, obtaining a series of relationship entity pair data from the data informationby using the relationship classification model, and storing the relationship entity pair as a basic knowledge carrier of the knowledge graph to complete the construction of the processing knowledge graph. By means of the mode, the information can be rapidly obtained, the working efficiency can be greatly improved, and knowledge provided by the knowledge graph is beneficial for better understanding information content and carrying out information public opinion analysis.

The invention relates to a city government hotline data integration platform comprising a unified basic support platform and the service application system of each data source department. The unified basic support platform comprises a basic facility layer, a data resource layer, an application support layer and an application service layer which are arranged from the bottom to the top in turn. The data resource layer acquires the data of all the data source departments. The application support layer performs basic arrangement on the data and acquires the basic data of the data resource layer and then performs the basic service and is used for providing service support to the application service layer. The service application systems establish information interaction channels with the application service layer and realize the application service for the data source departments. Various government affair service hotlines are integrated based on the mayor's open telephone, and the public service hotlines with integration of consulting, service, assistance, supervision and complaint, public opinion analysis and coordinated supervision are established so that various non-urgent appeals of the people put forward to the government apart from urgent alarm are received in a unified way.

The invention provides an enterprise public opinion analysis method and device based on self-attention. The method comprises the steps that text data are extracted from an obtained Chinese enterprisepublic opinion data set, emotion tags are checked, the text data are preprocessed, stop words, low-frequency words and missing values are removed, the processed text data are obtained, and the text data comprise a title, an abstract and a text of the public opinion data; lexical learning is performed on the processed text data by using a bidirectional GRU algorithm, performing training to obtain alanguage model, and learning is performed to obtain a corresponding word vector; the word vectors are input into a self-attention encoder to learn the relationship between words, and a syntactic-level text representation mode is learned by calculating a self-attention coefficient; and the probability of each sentiment category is calculated through an sentiment analysis classifier according to the syntactic-level text representation mode to obtain an analysis result of enterprise public opinions.

The invention relates to the public opinion analysis technology and discloses a sentiment classification system and method based on a support vector machine. The sentiment classification system and method are used for finding public opinions from user comment information fast and accurately. A crawler module is used for acquiring comment information published on a forum by a user, data is subjected to word segmentation and other preprocessing, a feature word group of a comment text and training data with typicality are obtained, then the training data is subjected to sentiment annotation, the support vector machine is utilized for calculating the training data, and a classification model is obtained; the comment file to be classified is analyzed through the classification model, and an expected sediment state is obtained; finally, a visualization module is utilized for showing the classification result, the user is helped to know user sediment based on different entity objects (keywords) fast, then the public opinion of the internet is known, and the system and method are suitable for public opinion analysis of websites and forums.

The invention relates to a big data engine technology, is applied to artificial intelligence, and provides an enterprise public opinion analysis method. The method comprises the steps of obtaining andpreprocessing a public opinion text; obtaining multiple phrases, identifying the named entity and determining the public opinion text of the identified enterprise entity as a target text; performingsubject recognition on the target text; obtaining identification results, classifying the target texts; obtaining classification results, inputting the target text, the recognition result and the classification result into a pre-trained emotion model, obtaining sentiment results, selecting a target text of which the emotion result is negative emotion, and using the early warning text as the earlywarning text; matching the early warning text with the early warning phrases in the configuration list, determining the matched early warning phrases as the early warning category of the early warningtext, obtaining the target enterprise of the early warning text, and forming the early warning information according to the early warning category and the early warning text, thus sending early warning out timely and accurately. The invention also relates to a blockchain technology, and the early warning information is stored in the blockchain.

A system capable of detecting and tracking a topic based on an opinion and a social influencer for each topic and a method thereof are provided. The system that detects and tracks the topic based on the opinion and the social influencer for each topic includes: a topic opinion analysis apparatus configured to extract an opinion for each topic from content on the Internet to generate a topic-opinion database (DB), and to extract a social influencer among social members ranked based on behavior characteristics to generate a topic-social influencer DB; a knowledge DB configured to store the topic-opinion DB and the topic-social influencer DB; and a topic detecting and tracking apparatus configured to detect an opinion on a designated topic requested to be searched based on the knowledge DB, search the social influencer of the designated topic, and collect and analyze content provided or generated by the searched social influencer.

The invention discloses an internet public opinion analyzing method and device. The internet public opinion analyzing method comprises the steps of obtaining target web pages of internet public opinion analysis from the internet, carrying out word segmentation on texts in the target web pages, obtaining text feature words of the target web pages from the result of word segmentation by utilizing the statistical algorithm, classifying the text feature words into positive text feature words and negative text feature words according to a positive and negative word dictionary practiced in advance, and judging whether the target web pages are the positive public opinion web pages or the negative public opinion web pages according to the classified results of the text feature words. The internet public opinion analyzing method and device improve the efficiency of analyzing tendency of internet public opinions.

The invention relates to a public opinion analysis method and system, computer equipment and a storage medium. The method comprises the following steps of: crawling public opinion information on a preset platform; determining an influence object type corresponding to the public opinion information; splitting the public opinion information to obtain a plurality of short texts; obtaining a public opinion factor corresponding to the influence object type and respectively extracting target keywords from the plurality of short texts according to the public opinion factor; and obtaining a public opinion analysis model corresponding to the influence object type, and inputting the extracted target keywords into the public opinion analysis model to obtain an emotional index corresponding to the public opinion information by calculation. By adoption of the method, the public opinion analysis efficiency and correctness can be improved.

The invention discloses a figure cooccurrence relation graph establishing method based on a specific behavior cooccurrence network. The method comprises the steps that firstly, a standard training data set and a triggering vocabulary are established, and all sentences to be tested are filtered with a string matching method to obtain a collection of candidate specific behavior sentences; then, whole word feature vectors of all the sentences in the standard training data set are extracted and used for training an SVM classifier to recognize specific behaviors, the candidate specific behavior sentences are classified with the trained SVM classifier, and specific behavior sentences are recognized; finally, a specific behavior cooccurrence network including named entities and the cooccurrence relation thereof is established, and a figure cooccurrence relation graph is established based on the specific behavior cooccurrence network. By means of the figure cooccurrence relation graph establishing method, the specific behavior sentences can be effectively recognized from a large quantity of real online public opinion data, the specific behavior cooccurrence network and the figure cooccurrence relation graph are established on the base, and a novel visualized analysis method is provided for public opinion analysis.

The invention discloses a time window and semantic meaning-based word variant normalization method and system. The method comprises the following steps of: 1) selecting corpuses in a set time period before an appearance time of a given variant in a social network according to the appearance time so as to form a candidate corpus set D1; 2) adding corpuses similar to the corpus where the word variant is located in the aspect of semantic meaning in the candidate corpus set D1 into a candidate corpus set D2; 3) extracting candidate words from the set D2 to obtain a candidate word set; and 4) calculating a score of each pair of candidate word and variant according to a literal similarity and a context feature similarity between each candidate word and the word variant, determining a candidate word corresponding to the word variant according to the calculation result, and taking the determined candidate word as a normalized word of the word variant. The system comprises an acquisition module, a filter module, and obtaining module and a normalized word obtaining module. According to the method and system, the texts of social networks become more normalized, and the public opinion analysis and hotspot time tracing are convenient.

The invention discloses a commodity futures news public opinion analysis method, which comprises the following steps of acquiring futures industry data, establishing a futures knowledge graph based onthe futures industry data, and the futures knowledge graph comprises an entity library and a relationship library; obtaining futures related public opinion news, performing label structuralization processing on the related public opinion news according to public opinion analysis dimensions, and outputting structuralized labels of the futures related public opinion news; wherein the structured label comprises a coarse-grained label mapped with an entity in an entity library; and obtaining an analysis request, and performing association analysis on the futures knowledge graph and the futures related public opinion news with the structured label according to the analysis request to obtain a corresponding analysis result. Public opinion analysis can be carried out at different levels, futuresrelated public opinion news and futures knowledge graphs are associated through the structured labels, systematic and systematic analysis on public opinion information of the futures market is achieved, and news public opinion analysis requirements of the futures industry are satisfied.

The invention relates to a search method of a real-time vertical search engine for a security industry. The search method comprises performing high-frequency directional fetching on news web pages through a server; performing formatting processing on news content of the fetched news web pages; performing evaluation calculation on the relevance of the formatted news content and relevant keywords and the influence on the public of the news content; storing results into a database and calculating weights of the search results according to multiple parameters and sorting and displaying the search results through a system when users search data. Accordingly, the passive synchronization of the search engine information and an information source can be achieved and the problems that the general search engine by the traditional search method is poor in timeliness and repeated in information are solved; the directional collection is only performed on an industry representative financial news release source of the Internet and accordingly the efficiency is high and the search results are timely and accurate; in addition, the search method is combined with a public opinion analysis technology and accordingly the search results can be sorted in multiple modes and the display effect is humanized.