44 results about "Emotional score" patented technology

A system and a method for generating an emotional profile of the user and deriving inference from the analytics of generated emotional profile is provided. The method involves sharing media content or online event in a connected environment; capturing user's reaction to the said content or event; generating an emotional score of the user to rate the media content or event; and sharing the emotional score within the connected environment.

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.

A system and a method for generating an emotional profile of the user and deriving inference from the analytics of generated emotional profile is provided. The method involves sharing media content or online event in a connected environment; capturing user's reaction to the said content or event; generating an emotional score of the user to rate the media content or event; and sharing the emotional score within the connected environment.

The invention discloses a machine learning-based method for analyzing the customer satisfaction degree of e-commerce products. The method comprises the steps of: obtaining e-commerce product review text and carrying out data preprocessing such as word segmentation, part-of-speech annotation and the like; selecting Chinese chunk mark symbols to manually annotate word segmenting results; training amodel on the basis of a Lib-SVM (Library for Support Vector Machine) tool, obtaining nominal Chinese chunks to serve as candidate product features and calculating TF-IDF (term frequency-inverse document frequency) value filtering features; constructing an emotional dictionary and calculating the emotional score of each feature of a product; training a word vector language model and obtaining vector representations of the product features; and carrying out customer satisfaction degree clustering on the product features on the basis of word vector similarity degree and calculating a total score.The method disclosed by the invention can be applied to product review text-based product recommending systems and has the beneficial effects that through the customer satisfaction degree analysis, five aspects of the product features are clustered and the product feature dimension and sparsity are reduced, so that the designed recommending systems have the performance of being faster and more accurate.

The invention discloses a GIF short video emotion recognition method and a GIF short video emotion recognition system fusing text information. The method firstly extracts the sequence features in theGIF short video by using a 3D convolutional neural network, and simultaneously extracts the image visual features in the sequence by using the convolutional neural network. Then the high-level semantic features are decoded by using convolution long-short memory recurrent neural network technology, and the probability distribution matrix of emotion classification is calculated, and the emotion scores of the video part are obtained by interval mapping. Then, the vocabulary which contains emotional information is selected from the words in the annotated text, and the emotional score is calculatedby the emotional scoring tool. Finally, the video emotion score and the text emotion score are given different weights to add and do the validity judgment, and the GIF short video emotion classification. The invention can effectively pay attention to the emotional information of images in the GIF video, simultaneously taking into account the temporal characteristics of the video stream, and fusesthe text information and the video information, thereby improving the accuracy and robustness of the emotional classification of the GIF video.

In the present embodiment, a system and a method for tagging a content based on individual cues, emotional score or emotional profile is provided, where the content is a video file, a webpage, a mobile application, a product review or product demo video. The method involves authorizing a user to access the content; capturing a user specific data, an application details and a content specific data in response to the content in real-time; analyzing the captured user specific data, the application detail and the content specific data to generate a user emotional profile; and tagging the user emotional profile with the content in a time granular manner.

The invention provides a music special effect matching method and device, and relates to the technical field of music recognition. The method comprises the steps: acquiring to-be-matched audio data, and detecting the beat of the audio data; determining whether the beat is a rhythm change point of the audio data or not: mark the beat as a starting point of an optional music passage if the beat is the rhythm change point of the audio data, or else generating a starting point list; selecting the starting point of the music passage in the starting point list according to a preset starting point selection rule, and dividing the audio data into multiple music passages; setting an emotional label of each music passage, and calculating the emotional score of each emotional label; searching for a visual expression effect matched with the emotional score in a pre-stored special effect database and storing the visual expression effect in association with the music passage, so that the audio datais read and the visual expression effect is played. The method and device provided by the invention facilitate the emotion interaction between the user and music, and improves the user experience.

A system and a method for generating an emotional profile of the user and deriving inference from the analytics of generated emotional profile is provided. The method involves sharing media content or online event in a connected environment; capturing user's reaction to the said content or event; generating an emotional score of the user to rate the media content or event; and sharing the emotional score within the connected environment.

The invention provides a voice early warning method, device and equipment and a storage medium, and relates to the technical field of data processing, which aim to analyze a voice emotion score and atext emotion score of each voice, fuse analysis results and monitor dynamic changes of emotions in real time. The method comprises the following steps: continuously acquiring each sentence of voice output by a voice acquisition device; converting each sentence of voice into text content; obtaining the voice feature of each sentence of voice and the text feature of the text content obtained by converting each sentence of voice; according to the voice features, judging the stability degree of the emotion of the user, and according to the text characteristics, judging the stability degree of theemotion of the user, and obtaining a comprehensive emotion score of each sentence of voice; and according to the change condition of the comprehensive emotion score along with the acquisition time ofeach sentence of voice, giving early warning.

A system and a method for tagging content based on individual cues, emotional score or emotional profile is provided, where the content is a video file, a webpage, a mobile application, a product review or product demo video is provided. The method involves authorizing a user to access the content; capturing a user specific data, an application details and a content specific data in response to the content in real-time; analyzing the captured user specific data, the application detail and the content specific data to generate a user emotional profile; and tagging the user emotional profile with the content in a time granular manner.

The invention provides a network public opinion analysis method, which comprises the following steps of: collecting the public opinion articles related to a first preset keyword, and performing word segmentation processing on the public opinion articles; matching a vocabulary set corresponding to each public opinion article with a plurality of second preset keywords, and labeling a first tag corresponding to the second preset keyword for each public opinion article; calculating an emotional score of the public opinion article, judging an emotional tendency of the public opinion article, and marking a second tag; respectively counting the total number of public opinion articles and the number of negative public opinion articles corresponding to each first tag; and respectively calculating the negative public opinion influence and the total public opinion influence corresponding to each first tag, and calculating the public opinion health degree of each first tag. The invention also provides an electronic device and a storage medium. According to the present invention, the monitoring and early warning ability of the public opinion can be improved, and the user is helped to make corresponding decision according to the public opinion event.

The invention discloses a method for intelligently analyzing a Chinese character emotional tendency through a computer. The method is characterized by comprising reading Chinese character paragraph files, segmenting the Chinese character paragraph files and performing word segmentation, part-of-speech tagging and syntactic interdependent relationship marking on segmentations to form extensible markup language (EML) files; reading the EML files, going through sentence extraction syntactic interdependent relationship pairs and assigning extracted words based on a dictionary, wherein a word in a positive word dictionary is assigned with 1, and a word in a negative word dictionary is assigned with -1; degree adverbs are divided into 5 grades according to degrees and assigned with 1.8, 1.5, 1.2, 0.9 and 0.5 respectively; and negative adverbs can be divided into -1 and -1.5 grades according to negative degrees; and going through the dictionary according to a formula: emotional score= negative words* adverb sum* adjectives and obtaining the emotional score of the Chinese character paragraph files; and the emotional tendency of the Chinese character paragraph files is judged according to the emotional score.

The invention provides a bullet screen distribution method and device, wherein the method comprises the steps of: S1, grouping users in a current live broadcasting room, and, according to word vectors of bullet screens to be distributed currently sent by the various users and word vectors of first historical bullet screens sent by the various users within the first pre-set historical time slot, obtaining the emotional scores of the bullet screens to be distributed and the first historical bullet screens by using a trained LSTM neural network; S2, according to the emotional scores of the first historical bullet screens, obtaining the total emotional scores of the first historical bullet screens sent by each group of users; and S3, based on judgement on the total emotional scores of the first historical bullet screens sent by each group of users, sending the bullet screens to be distributed to each group of users correspondingly according to the emotional scores of the bullet screens to be distributed. According to the bullet screen distribution method and device provided by the invention, the bullet screens are distributed according to the total emotional scores of the first historical bullet screens sent by each group of users within the first pre-set historical time slot and the emotional scores of the bullet screens to be distributed currently sent by the users; therefore, the bullet screens are directionally distributed to each group of users according to the emotional scores of the bullet screens; and the interaction participation enthusiasm of the users is improved.

The invention discloses an industry text emotion obtaining method, device and a storage medium, which relate to the technical field of natural language processing big data analysis. The method includes carrying out emotion sentence recognition on clauses in combination with an emotion knowledge base of a target field so as to form an emotion sentence set; according to the constructed affective dependency tree, determining a method of obtaining the scores of affective dependency; the long sentences are segmented into clauses, and the emotional scores of clauses, long sentences and paragraphs are obtained according to the method of obtaining the emotional dependency scores of the target text. Through the practical application test, the industry text emotion calculation method provided by theinvention can better reflect the industry text emotion, and the accuracy rate reaches 85% in the calculation of the long text chapter level emotion. Aiming at the texts of different fields, the invention can quickly construct the domain emotion thesaurus, and realize the fast self-adaptation of the emotion calculation of different industries.

The invention relates to a short text emotional value calculation method based on sentence structures and contexts. The method comprises: obtaining any to-be-analyzed text data; performing word segmentation processing on the text data through a Java-based word segmentation program; combined with a string matching algorithm, determining sentence structures of the to-be-analyzed text by computer programming; and making the to-be-analyzed text corresponding to the corresponding sentence structure, calculating an emotional score of each clause; implementing a contextual emotional score of the to-be-analyzed text. Part of the emotional score is an emotional value from comment data appearing before the text, influence on current data is calculated according to the distance from the current data,and the other part is an emotional value produced by the news. Establishment of an emotional dictionary is based on an open-source emotional dictionary, that is, emotional vocabulary noumenon of Dalian University of Technology. The emotional value of the short text is calculated combined with the emotional dictionary. Accuracy of the emotional values calculated by the method is higher, and effectin the field of public opinion analysis will be better.

The invention discloses an online education auxiliary scoring method, which comprises the steps of obtaining pictures of students in class according to a camera, periodically detecting whether students appear in pictures or not through a face detection algorithm, and counting to acquire the number of times of the students appearing in the pictures so as to judge whether each student listens to class or not and whether the students focus attention or not in class or not to acquire attention scores; capturing videos of the students during teacher-student interaction, performing classification modeling training on student expressions based on an action classification method of a video sequence, identifying emotion categories of the students to acquire emotion scores; and calculating final education auxiliary scores according to the weights of the attention scores and the emotion scores. The invention is mainly applied to teaching, scores the attention and emotion mining of student subjects, and accurately reflects the listening quality of the students by scores.

The embodiment of the invention provides a method for calculating the emotional score and related equipment, which is used for solving the problem of collecting and summarizing the functional feedbackmanually collected by the user in the existing live broadcast platform, realizing automatic identification of the functional problem and giving the emotional bias score of each function, and facilitating the product optimization judgment. The method comprises the following steps: obtaining user feedback text of the live broadcast platform; Segmentation of the user feedback text according to a preset rule to obtain a segmentation set; Determining a functional word set according to the word segmentation word set; Matching each functional word in the functional word set with a functional word set in a vocabulary set corresponding to the live broadcast platform to determine a functional sentence set; Determining a target short sentence set according to the functional sentence set; Determiningan emotion score synthesis value of the target short sentence set; Determining an emotion score of the target function according to the emotion score composite value.

The embodiment of the invention provides an identification method and an identification device for user intention. The method comprises the following steps: carrying out emotional analysis on voice information of users to obtain target emotional scores corresponding to the voice information; judging whether the target emotional scores are lower than a preset emotional score threshold; if the target emotional scores are lower than the preset emotional score threshold, setting an unclarified time threshold according to the target emotional scores; determining the times of unclarified intention in the communication with the user according to the voice information; if the determined times of the unclarified intention are lower than a time threshold of the unclarified, starting a self-service system to identify the user intention. The method provided by the embodiment of the invention has the advantages that the times of the unclarified intention can be dynamically set, service is intelligently provided for the users and user problems are timely solved, so that user time is saved and user complaints are reduced.

The invention provides a contacts sorting method and device. The method can comprise the following steps: acquiring emotional scores of various contacts in an address book; and sorting the contacts inthe address book according to the emotional scores of various contact. The embodiment of the invention can quickly find the contacts needed to be searched.

The invention provides an emotion analysis method and apparatus, and a storage medium. The method comprises the steps of obtaining a target entity in a title of a to-be-detected article; obtaining a target statement containing a target entity in the text of the to-be-detected article; according to an emotion analysis model, obtaining the title and an emotion score of a target entity in the target statement, wherein the emotion analysis model is used for representing the corresponding relation between the entity and the emotion score in the statement; and according to the title and the emotional score of the target entity in the target statement, obtaining the emotional score of the target entity in the to-be-detected article. The emotion score of the target entity can be obtained from the article containing multiple entities, and the credibility of an article emotion analysis result is improved.

The invention discloses a method and system for emotional recognition of GIF short videos fused with text information. The method first uses 3D convolutional neural network to extract sequence features in GIF short videos, and at the same time uses convolutional neural network to extract visual features of images in the sequence ; and then use the convolutional long-short-term memory recurrent neural network technology to decode the high-level semantic features after the fusion of the two, calculate the emotion classification probability distribution matrix, and perform interval mapping to obtain the emotional score of the video part. Then, the words containing emotional information are screened out from the words in the annotation text, and the emotional score of the text is calculated using the emotional scoring tool. Finally, the emotional score of the video and the emotional score of the text are added with different weights and the effectiveness is judged to classify the sentiment of the GIF short video. The invention can effectively pay attention to the emotional information of the image in the GIF video, and at the same time take into account the timing characteristics of the video stream, and integrate the text information and the video information, thereby improving the accuracy and robustness of the emotional classification of the GIF video.

The invention relates to machine learning in artificial intelligence, and provides a resource sharing method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining resource request face pictures which are uploaded by at least two resource requesters and are photographed according to reference face pictures published by resource publishers; performing feature extraction on the resource request face picture to obtain resource request micro-expression features, and inputting the resource request micro-expression features into a trained micro-expressionanalysis model for analysis to obtain a resource request emotion score corresponding to the resource requester under each candidate emotion state type; performing feature extraction on the reference face picture to obtain reference micro-expression features; inputting the reference micro-expression features into a trained micro-expression analysis model for analysis; and obtaining a reference emotional state type corresponding to the resource publisher, and obtaining a target resource request emotional score corresponding to each resource requester under the reference emotional state type, thereby performing resource segmentation on to-be-shared resources of the resource publisher, and determining a resource sharing result of each resource requester.

The invention discloses a social text sentiment classification method and system based on a text graph neural network, and belongs to the technical field of natural language processing. Comprising the following steps: receiving a target text, and removing an abnormal value in the received text; obtaining word embedding of the target text by utilizing a BERT model; obtaining emotional polarity characteristics of the target text, calculating an emotional score of each word of the target text by using a SentiWordnet emotional dictionary source, and taking a final score of each word as the emotional polarity characteristics of the word; splicing the word embedding and sentiment polarity features of the target text to form an initial word vector; and constructing the target text into a text graph structure, taking the initial word vector as a node initial feature of a text graph, then performing feature extraction by using a text graph neural network message passing mechanism, and finally performing sentiment classification on the extracted feature. The method not only considers the context features in the speech, but also considers the mutual relation between the speech, so that the sentiment classification is more accurate.

The invention discloses a text sentiment recognition method and device, a storage medium and computer equipment, relates to the technical field of information, and mainly aims to determine sentiment tendencies of a text under different sentiment analysis granularities and meet the requirements of a user for the sentiment tendencies of the text under the different sentiment analysis granularities.The method comprises the steps of splitting a to-be-recognized text into statements; utilizing a preset emotion score algorithm corresponding to different emotion analysis granularities to calculate an emotion score of each statement under different emotion analysis granularities; and determining the emotional tendency of the to-be-recognized text under different emotional analysis granularities based on the calculated emotional score of each statement under different emotional analysis granularities. The method is suitable for text sentiment recognition.

The invention discloses a method for automatically generating a novel text emotion curve and carrying out prediction recommendation. The emotion curve generated by the invention can more accurately reflect a text emotion change situation. A relationship between the novel text emotion curves can be innovatively utilized to predict novel statistics, and provided downloading amount prediction exhibits higher positive relevance. The recommendation of a related novel text provides a brand new angle for a relevant recommendation task. The method comprises the following main steps that: after a novel text corpus is preprocessed, obtaining the word list of the novel, calculating emotional scores in sequence through a text window, and carrying out converging to obtain the emotion curve; through the emotion curve, calculating a dynamic time regular distance matrix between every two texts; for the dynamic time regular distance matrix, using an improved Gaussian process to give the downloading amount prediction; and according to the dynamic time regular distance matrix, giving relevant text recommendation.

The invention provides a music special effect matching method and device, and relates to the technical field of music recognition. The method comprises the steps: acquiring to-be-matched audio data, and detecting the beat of the audio data; determining whether the beat is a rhythm change point of the audio data or not: mark the beat as a starting point of an optional music passage if the beat is the rhythm change point of the audio data, or else generating a starting point list; selecting the starting point of the music passage in the starting point list according to a preset starting point selection rule, and dividing the audio data into multiple music passages; setting an emotional label of each music passage, and calculating the emotional score of each emotional label; searching for a visual expression effect matched with the emotional score in a pre-stored special effect database and storing the visual expression effect in association with the music passage, so that the audio datais read and the visual expression effect is played. The method and device provided by the invention facilitate the emotion interaction between the user and music, and improves the user experience.

The invention discloses a multi-modal emotion analysis method based on a multi-dimensional attention fusion network, which includes: extracting voice preprocessing features, video preprocessing features, text Preprocessing features; then construct the multi-dimensional attention fusion network for each modality, use the autocorrelation feature extraction module inside the network to extract the first-level autocorrelation feature and the second-level autocorrelation feature, and then combine the three modalities Combining autocorrelation information, using the cross-modal fusion module inside the network to obtain the cross-modal fusion features of the three modalities; then using the second-level autocorrelation features and cross-modal fusion features to combine to obtain modal multi-dimensional features ;Finally, splicing the multi-dimensional features of the modalities, determining the sentiment score, and performing sentiment analysis; sentiment analysis.

In order to alleviate the influence of word spelling mistakes, wrong order and mixed emotions on the final grasp of the overall sentiment score of reviews, these noise and mixed emotions can be dealt with in a targeted and orderly manner. The present invention designs a comprehensive deep capsule network classification model , simulating the logical steps of human reading, by capturing feature information at the word level, phrase level, and sentence level, respectively, to model reviews, specifically, modeling at the word level and phrase level with misspelling errors and word order errors Corresponding to the noise problem, the modeling at the sentence level corresponds to the emotional mixing problem, that is, each short sentence is regarded as a synonymy, and the impact of different synonymy groups on the final overall emotional attitude is dynamically considered. In terms of implementation, the BERT WordPiece vector and convolution can be used as word-level and phrase-level features, and then the capsule network can be used to obtain the final vector representation at the sentence level for classification.