63 results about "Sentiment score" patented technology

A method for managing an organization's performance includes: receiving an adjacency matrix (AM) from a document attention module; determining, by employing a first model, a first k-hop value for a first transcript and a second k-hop value for a second transcript based on the AM; analyzing the first k-hop value, the second k-hop value, and the AM; generating, by employing a second model, a second graph representation based on relationships in the AM; transforming the second graph representation into embedding vectors (EVs); analyzing, by employing a language model, the EVs to extract features; analyzing the features to generate a first sentiment score (SS) for a first sentence of the first transcript and a second SS for a second sentence of the first transcript; and classifying the first sentence as a first negative sentence and the second sentence as a second negative sentence.

The embodiment of the invention provides a marketing verbal skill generation method and device, equipment, a storage medium and a program product, and relates to the field of financial science and technology and artificial intelligence. The method comprises the following steps: acquiring multi-modal data; performing cross-modal feature fusion on the voice data and the text data to generate a joint feature vector; based on the joint feature vector, through an attention mechanism, a multi-dimensional emotion evaluation result is generated, and the multi-dimensional emotion evaluation result comprises an evaluation result of at least one preset emotion dimension; and according to the multi-dimensional emotion evaluation result, generating and pushing a marketing verbal skill corresponding to the multi-dimensional emotion evaluation result. According to the method provided by the invention, through joint modeling of voice and text features and in combination with an attention mechanism, a key emotion region is focused, so that the accuracy of emotion scoring is remarkably improved, and the attitude of a customer can be reflected more truly; according to the emotion evaluation result, the optimization suggestion is generated, the generated verbal skill accurately matches the demand of the customer, the response efficiency of the marketing verbal skill is improved, and the customer emotion improvement rate is greatly improved.

The invention discloses a data behavior prevention and control and anomaly detection system based on big data, and relates to the technical field of data security. Comprising a data acquisition module used for acquiring user behavior data; the feature construction module is used for constructing emotion features, space-time trajectory features and abnormal data features according to the preprocessed user behavior data; the calculation module is used for calculating to obtain a comprehensive score according to the emotion features, the spatial-temporal trajectory features and the abnormal data features; the threshold judgment module is used for respectively carrying out threshold judgment on the comprehensive emotion score, the comprehensive space-time score and the comprehensive generated data score, and generating a first abnormal signal, a second abnormal signal and a third abnormal signal when a threshold condition is exceeded; and the model construction module is used for constructing an anomaly detection model and generating a fourth anomaly signal when an anomaly is detected. Through multi-dimensional feature fusion, real-time anomaly detection and a deep learning technology based on big data, the accuracy of abnormal behavior recognition is improved, and the security and reliability of the system are enhanced.

Artificial Intelligence driven system for managing deployment actions based on sentiment analysis includes performing operations including obtaining an assigned feature name for an application feature of a target application, selecting, using respective vector encodings of an application feature name that include the assigned feature name, and filtering a collection of feedback using a set of application feature names of an application feature to obtain a filtered set of feedback. The operations further include generating a sentiment request prompt requesting a sentiment score and comprising the filtered set of feedback and the set of application feature names, and processing, by a large language model (LLM), the filtered set of feedback with the set of application feature names to generate the sentiment score for the application feature. The operations further include selecting a recommended action based on the sentiment score and updating the target application according to the recommended action.

This application discloses a method and apparatus for quantifying the quality of rural hotels, as well as a method for processing rural hotel information, belonging to the field of data processing and information technology. The method includes: screening and confirming rural hotels from accommodation facilities within a target area according to preset identification rules to obtain a rural hotel directory; collecting online user review text data for each rural hotel; extracting high-frequency feature words based on the online user review text data, performing semantic analysis, and constructing a quality evaluation index system; mapping each review text to the corresponding index in the quality evaluation index system, and obtaining the global weight of each level of index by statistically analyzing the number of review texts corresponding to the first-level and second-level indicators; performing sentiment analysis on each review text and calculating the sentiment score of each level of index; and calculating the quality quantification score of each rural hotel based on the sentiment score corresponding to each level of index and its corresponding global weight. The embodiments of this application can improve the accuracy of rural hotel quality quantification.

A method for managing an organization's performance includes: receiving an adjacency matrix (AM) from a document attention module; determining, by employing a first model, a first k-hop value for a first transcript and a second k-hop value for a second transcript based on the AM; analyzing the first k-hop value, the second k-hop value, and the AM; generating, by employing a second model, a second graph representation based on relationships in the AM; transforming the second graph representation into embedding vectors (EVs); analyzing, by employing a language model, the EVs to extract features; analyzing the features to generate a first sentiment score (SS) for a first sentence of the first transcript and a second SS for a second sentence of the first transcript; and classifying the first sentence as a first negative sentence and the second sentence as a second negative sentence.

The application discloses an intelligent customer service interaction method and system, and relates to the technical field of intelligent customer service interaction.The method comprises the following steps: obtaining a user request; calculating a comprehensive sentiment score through multi-modal sentiment analysis based on the user request; classifying the user's emotional state according to the comprehensive sentiment score, initializing the dialogue context, and constructing a dialogue context graph according to the classification result; dynamically adjusting the response mode of the intelligent customer service according to the emotional state and the user's question in the dialogue context graph, and collecting the feedback information of the user; and dynamically adjusting the dialogue strategy based on the feedback information of the user and the dialogue context.The application can comprehensively capture the emotional state of the user from multiple angles such as text, voice and facial expression through multi-modal sentiment analysis, improves the accuracy and robustness of emotion recognition, provides a reliable basis for subsequent dialogue management and response mode, and improves the experience satisfaction and service quality of the user.

The invention discloses a social media data fused urban inefficient land intelligent identification method, which comprises the following steps of: realizing multi-source data space fusion by acquiring a social media geographic marking text, an urban updating policy file, a remote sensing image and basic geographic data; a natural language processing and topic modeling method is used for extracting keywords from the city update text, a topic-feature word mapping table is constructed, a RoBERTa-l arge model and a semantic rule are combined to calculate a text emotion score, and a land use negative emotion map is generated; and constructing a multi-dimensional evaluation system including indexes such as plot ratio and road network density, calculating a composite index by adopting an AMP algorithm, dynamically allocating weights through an entropy weight method, and judging the low-efficiency land. And a white list area is introduced for verification and parameter iterative optimization, and finally, an identification result is visually displayed in WebG IS and AR platforms, so that the accuracy and decision support capability of urban low-efficiency land identification are improved.

The application discloses a kind of text writing revision auxiliary method and system, it is related to artificial intelligence technical field, method includes: obtaining the current text description of user, extract entity text and carry out semantic sorting, generate entity text sequence;The specification score, sentiment score and writing object level of each entity text are identified by word vector model, generate first projection point;With each entity text as center text, the projection point of adjacent text in window is fused, to obtain second projection point, build current scene polyhedron;Each first projection point of entity text sequence and current scene polyhedron are input to writing detection model, model is encoded by multi-head self-attention and 3D grid convolution operation, generate text and scene vector and weighted fusion, decode output error prediction and position push to user.By implementing the application, the multidimensional characteristics of the scene are retained, thereby solving the problem that a single scene tag cannot reflect the real scene with multiple elements interwoven in the prior art.

The invention discloses a method, device and equipment for adjusting background light of a terminal with a screen and a storage medium, and relates to the technical field of intelligent hardware. The method comprises the following steps: acquiring figure video data, text data and audio data in a video; inputting the character video data, the text data and the audio data into a sentiment analysis model, extracting modal features and inter-modal features of the character video data, the text data and the audio data through the sentiment analysis model, and calculating decoupling feature input according to a relation among the modal features, the inter-modal features and the decoupling features; outputting a target sentiment score predicted value according to a mapping relationship between the decoupling features and the sentiment score predicted value; and adjusting the background light according to the target emotion score predicted value. According to the embodiment of the invention, the real-time adaptive adjustment of the background light is realized, and the background light interaction adjustment capability is improved.

The invention discloses a book recommendation method fusing existing user borrowing behaviors of a library and Internet resources. According to the method, book quality evaluation indexes comprehensively consider the following four factors: influence weights of readers on book types, return timeliness, borrowing group breadth and Hubamp; an Authority method is used for comprehensively measuring the quality of the books; the implementation process of the recommended purchase method comprises the following steps: firstly, collecting all recommended purchase requests to be processed from a library database, and constructing an initial recommended purchase list; thirdly, screening first three books which are most similar to the books in the list through cosine similarity, and extracting borrowing data of the first three books; thirdly, combining the borrowing records with comments of read users on the social media, and performing multi-modal sentiment analysis to obtain sentiment scores; then, integrating the emotion score with the book special information by using a random forest model to form a comprehensive score; and finally, optimizing recommended purchase list sorting according to the comprehensive score, and determining a final book purchase priority.

The invention discloses an interactive audience participation feedback method based on sentiment analysis and behavior prediction. User behavior data and text sentiment information are collected in real time through an audience terminal, behavior analysis and sentiment recognition are executed in parallel on a cloud server, and behavior tags and multi-level sentiment scores are generated; in combination with the historical portrait of the user, personalized interactive feedback content including an expression package, a knowledge point card, a microphone connection invitation or a voting entry is dynamically generated through a weighted scoring mechanism, and the pushing frequency is controlled to be not more than once every three minutes to avoid interference; meanwhile, when the individual continuously carries out negative feedback or the overall negative emotion ratio of the live broadcast room exceeds a threshold value, an instruction is automatically sent to the live broadcast stream pushing end, and real-time content adjustment is achieved. The method effectively solves the problems that traditional live broadcast interaction lags behind, is single in form and lacks emotional response, remarkably improves the audience immersion, participation and platform retention rate, and is suitable for various live broadcast scenes such as education, entertainment and e-commerce.

A method and system for investing in securities, equities, and funds is currently provided. The method comprises a client selecting at least one security, equity, or fund from their current holdings, comparing their selection with other available securities, equities, and funds in a database and calculating a sentiment score for each of the compared alternatives. The sentiment score is a calculated likeness between, or an inverse distance from, the client's selection and the compared securities, equities, or funds. The compared securities, equities, or funds are ranked by their sentiment scores and an expense ratio, a 12b-1 fee, and a front load fee associated with at least an upper ranked portion of the compared securities, equities, or funds is reported.

The invention discloses a comment emotion weighting mechanism-based fine-grained score graph recommendation method, which comprises the following steps of: taking users and items as graph nodes, taking continuous scores as edge weights, taking comment semantic features as edge features, carrying out message transmission and aggregation through a graph convolutional network, and realizing score prediction in combination with multi-source representation fusion. A large language model is used for analyzing comment emotion to generate an emotion score, the emotion score and an original score of a user are subjected to dynamic weighted fusion, an emotion weighted score is generated, rough expression of traditional discrete scores is broken through, user preferences are described more finely, and the problem that the score is inconsistent with the comment emotion is relieved. Meanwhile, in order to fully mine multi-layer preference information in comments, a three-layer attention mechanism is introduced by research, and semantic information is extracted and aggregated layer by layer from a word level, a sentence level and a comment level. Experimental results of comparing three data sets in Amazon show that compared with an optimal comparison method, the mean square error of the recommendation method is increased by about 2.87%, 2.08% and 2.17%.

The present application relates to the technical field of data information processing, in particular to a financial text sentiment feature optimization method and system based on multi-supervision signal LoRA fine-tuning. The method comprises: obtaining financial text and transaction data; generating at least three groups of sentiment scores based on a large language model, including original scores, first fine-tuned scores based on financial special model label fine-tuning, and second fine-tuned scores based on future yield label fine-tuning; calibrating each version of the scores and constructing multi-dimensional emotion features; fusing each version of the emotion features with technical index features to form multiple candidate feature sets; using time series cross-validation and multiple heterogeneous evaluation models to systematically evaluate and statistically compare the candidate feature sets, and selecting the optimal emotion feature source according to the performance; and deploying a prediction model based on the optimization result. The present application realizes empirical comparison of fine-tuning strategies and objective optimization of feature sources, can avoid invalid fine-tuning overhead, and has a standardized engineering implementation process.

An information handling system includes a hardware processor to execute computer-readable program code instructions of an AI productivity tool software module to process an initial user query input received from a user and correlate the user query input to a capability intent action and to initiate the capability intent action to respond to the user query input, and execute a sentiment analysis software application to analyze a subsequent user query input and generate feedback sentiment data describing a positive or negative sentiment of the subsequent user query input in a generated transaction log including the initial user query input, the responsive capability intent action, the subsequent user query input and a feedback sentiment score of the subsequent use query input that is transmitted to a remote policy management server for an intent technology decision maker to generate capability intent action policies based on the feedback sentiment data.

The application provides a regional data monitoring method and system, wherein the method comprises the following steps: performing semantic extraction on a financial text to obtain an initial semantic vector of each segmented word in the financial text; determining an emotional segmented word and an emotional score of the emotional segmented word in the financial text based on an emotional dictionary, and determining a trend expression word and a trend score of the trend expression word in the financial text based on a trend expression dictionary; strengthening the initial semantic vector of each emotional segmented word based on the emotional score of the emotional segmented word, and strengthening the initial semantic vector of each trend expression word based on the trend score of the trend expression word, to obtain a strengthened semantic vector of the emotional segmented word and the trend expression word; performing regression analysis on the strengthened semantic vector of the emotional segmented word and the trend expression word and the initial semantic vector of the remaining segmented word in the financial text to obtain an emotional score of the financial text; and based on the emotional score of the financial text, statistical target regional financial public opinion data. The application realizes more accurate public opinion data monitoring of a target region.

A system is provided.SOLUTION: A system comprising: means for collecting in-house text-based communication data; means for feeding the collected data into a natural language processing engine and performing sentiment analysis; means for storing the analysis results and calculating a sentiment score; means for generating and notifying an alert when the sentiment score falls below a set threshold; and means for generating and displaying periodic reports based on the sentiment score and analysis results.SELECTED DRAWING: Figure 1

Machine learning technology can analyze in real-time the data from a call between a person and a customer service representative. Based on this analysis, a server can determine a sentiment score that describes a sentiment expressed by the person or the customer service representative. If the server determines that the sentiment score is less than or equal to a pre-determined value, the server can inform the customer service representative's manager so that the manager can take further action to help the person and / or the customer service representative.

Embodiments are directed to computer-implemented methods of governance risk analysis and scoring of documents that includes: extracting raw text from documents; transforming individual words from the raw text into nodes and edges that create networks of nodes and edges; analyzing certain words pertaining to governance, and graphing the words to a network of nodes and edges; assessing the certain words in the network using mathematics, graph theory, or NLP; determining a semantic score for each node using the number and weight of edges surrounding each node; determining a saliency score based on a word's reoccurrence in the raw text; determining a risk sentiment score determined by the seed dictionary with the list of words exhibiting risk; determining a node score for a certain node based on the semantic score, the saliency score, and the risk sentiment score; and iterating the scores automatically every time a new filing is uploaded.

The present application relates to the technical field of information technology monitoring, and more particularly to an information technology data monitoring method and system based on deep learning. The method comprises the following steps: S1: obtaining multi-modal signals and metadata, and obtaining a multi-modal comprehensive sentiment score, a decision pressure metric and a confidence estimate according to the multi-modal signals and metadata; S2: constructing a sentiment trajectory sequence under each operation step of the user according to the multi-modal comprehensive sentiment score; and S3: monitoring a stress-confidence paradox, emotional disconnection, cross-cultural baseline deviation and commitment-fulfillment consistency deviation based on the decision pressure metric, the confidence estimate and the multi-modal comprehensive sentiment score within a sliding window. The present application can improve the accuracy of monitoring by monitoring the stress-confidence paradox, emotional disconnection, cross-cultural baseline deviation and commitment-fulfillment consistency deviation, and can balance sensitivity and false alarm control by implementing an anonymous or real-name hierarchical strategy.

The invention discloses a vehicle machine voice data screening method and device, a medium and a program product, and relates to the technical field of vehicle machine interaction, and the method comprises the steps: obtaining historical voice interaction data which comprises a user instruction text, a vehicle response text and a user feedback text, the vehicle response text is obtained by responding to the user instruction text by the vehicle, and the user feedback text is obtained by feeding back to the vehicle response text by the user; performing emotion scoring on the user feedback text based on the emotion analysis model to obtain an emotion scoring result, and determining a negative feedback text in the user feedback text based on the emotion scoring result; and screening target voice interaction data associated with the negative feedback text from the historical voice interaction data. The defects that the coverage rate of manual screening is low, and negative feedback of the user cannot be comprehensively captured are overcome.

The application relates to a car evaluation sentiment analysis method and device, electronic equipment and a storage medium, wherein the method comprises the following steps: crawling a plurality of text comment information of a vehicle; calculating a comment sentiment score of each piece of text comment information of the plurality of text comment information; and inputting the comment sentiment score into a pre-trained car evaluation sentiment analysis model to output an actual evaluation sentiment of the vehicle based on the front and rear correlation information and key features of the text. The car evaluation sentiment analysis model is obtained by combining a long short-term memory network (LSTM) and a text classification (Text CNN) neural network. According to the application, the car evaluation sentiment analysis model is obtained by combining the long short-term memory network (LSTM) and the text classification (Text CNN) neural network, the plurality of text comment information of the vehicle is used as a training set, and the car evaluation sentiment analysis model is trained, so that the sentiment analysis of the car evaluation is realized, the analysis efficiency is improved, and the sentiment analysis result is more professional and targeted.

The invention relates to the technical field of network public opinion early warning, in particular to a network public opinion early warning method and device. The method comprises the steps of obtaining to-be-detected text data; based on the to-be-detected text data, determining a high-dimensional semantic embedding vector and an emotion score of the to-be-detected text data; based on the emotion score and the high-dimensional semantic embedding vector, determining a semantic emotion composite embedding vector; determining attenuation similarity of the semantic emotion composite embedded vector based on the semantic emotion composite embedded vector and a historical vector corresponding to a historical text in any node in a preset network; wherein the preset network comprises a plurality of nodes; and when the attenuation similarity is greater than the first preset threshold, sending the first early warning information to the preset terminal, so that the method can accurately perform early warning on the network public opinion.

Systems including one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform operations: determining web text sentiment scores for web texts; creating a ranked list of one or more match words; scoring the one or more match words in the ranked list; labeling, using a generative model, the one or more match words to create labeled training data; training, using the labeled training data, a word-based classifier to identify non-conforming web text; receiving first web text for display on the website; using the word-based classifier, as trained, to identify the first web text as non-conforming web text; and automatically not displaying the first web text on the website when at least the classifier score exceeds the predetermined threshold. Other embodiments are described.

The application discloses a social text stand identification method and system based on claim generation and consistency analysis, relates to the field of natural language processing, and comprises the following steps: obtaining original user tweet texts of users and preprocessing the original user tweet texts to obtain preprocessed user tweet texts; constructing text pairs according to the preprocessed user tweet texts; inputting the text pairs into a pre-trained semantic consistency identification model to obtain consistency identification results; wherein the consistency identification results comprise semantic consistency scores and semantic stand classification results; performing sentiment analysis on the preprocessed user tweet texts based on a pre-set sentiment dictionary, and calculating sentiment tendency scores; weighting and fusing the semantic consistency scores and the sentiment scores to obtain comprehensive stand tendency scores; and obtaining the stands of the users according to the comprehensive stand tendency scores and the semantic stand classification results. The application improves the stability and reliability of social text stand identification in scenes with strong emotions and variable language styles.

The application discloses an e-commerce comment noise reduction method and device, medium and product, and relates to the field of data analysis. First, the original data of user comments is obtained, including user ID, comment ID, comment time, comment content (including comment text and comment pictures), IP attribution of the comment user and merchant information; a time decay factor is calculated according to the comment time; text processing and sentiment analysis are performed on the comment text to calculate a good / bad sentiment score; a picture-text correlation check is performed on the comment pictures and the comment text to calculate a picture-text score; a spatial correlation check is performed according to the comment pictures and the IP attribution of the comment user to calculate a geographic score; an industry coefficient is determined according to the merchant information; a credibility score is calculated according to the time decay factor, the good / bad sentiment score, the picture-text score, the geographic score and the industry coefficient; and valid / invalid comments are identified according to the credibility score, which greatly improves the efficiency and accuracy of valid / invalid comment identification.

A computing device can determine a negative sentiment score based at least in part on user posts to a social media platform within a predefined interval of time. Then, the computing device can compare the negative sentiment score to a moving average sentiment score to determine whether the negative sentiment score exceeds a threshold value based at least in part on the moving average sentiment score. The computing device can then generate an alert in response to a determination that the negative sentiment score exceeds the moving average sentiment score by the predefined amount. Finally, the alert can be formatted according to a recipient defined format and transmitted to a client device, thereby activating a client application to cause the client application to display the alert and enable the client device to connect to the computing device to view additional information about the alert.

The invention provides a theme-based automobile public praise data analysis method, system and device and a medium, and relates to the technical field of public praise data analysis. According to the method, aggregated word-of-mouth text data related to a specific automobile theme is collected and preprocessed to generate structured text data. Secondly, converting the data into text feature vectors by adopting embedded representation learning, and constructing a topic distribution diagram containing semantic association, affiliation and time association on the basis of the text feature vectors; and finally, fusing information of the theme distribution map and the text feature vector by using a deep learning model, and outputting an emotion score for a specific automobile theme. The method can effectively capture and analyze emotional tendencies in a large amount of user generated contents, and provides more accurate and comprehensive emotional evaluation for a specific automobile theme.

The application provides a car brand recommendation method and system based on an attention mechanism and an LSTM, and relates to the technical field of data processing. User comment data is input into a sentiment analysis model, is processed by an LSTM layer and an attention layer in the sentiment analysis model, and the predicted sentiment scores of the user on multiple aspects are obtained. According to the predicted sentiment scores, the importance scores of each aspect, the sentiment demand scores of each aspect and the overall feature vector of the car are obtained, and then the preference degree of the user in each aspect is obtained. According to the preference degree of the user, the overall feature vector of the car and a random forest classifier, the recommended car brand is obtained. The LSTM layer and the attention layer in the sentiment analysis model process the user comment data, provide deeper text understanding and reduce the required amount of user comment data. The correlation between the personal preference of the user and multiple feature dimensions of the car brand is established, and the accuracy of the recommendation is improved.