Sentiment-Aware Search Query Parsing and Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search technologies fail to provide users with sentiment-aware results directly from sentiment-aware search queries, requiring manual efforts to extract sentiment insights from lexical search results, as they lack mechanisms for explicit sentiment analysis and filtering based on user-defined sentiment interests.
Innovation Solution
A method and system that parse search queries to isolate lexical terms and sentiment analysis components, allowing for sentiment assessment and filtering of documents, using entity extractors and classifier models to generate sentiment-aware results that can include positive or negative indications or scores, even for concepts not explicitly mentioned in the query.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual effort is used to extract sentiment insights from lexical search results, then users can obtain sentiment information, but the process is time-consuming and inefficient
Solution Approach 1:
The system enables self-service by automatically performing sentiment analysis on retrieved documents without requiring user manual extraction. The sentiment analysis component processes documents autonomously, returning sentiment-aware results that directly provide the desired insight while eliminating the time-consuming manual extraction process.
Solution Approach 2:
The patent replaces the mechanical manual extraction process with an automated computational sentiment analysis system. Instead of users manually reading and extracting sentiment information, the system uses natural language processing and classification algorithms to automatically identify and extract sentiment insights from documents.
2Productivity
If lexical search is performed without sentiment analysis, then search results are returned quickly, but users cannot directly obtain sentiment-aware results
Solution Approach 1:
The system merges lexical search and sentiment analysis into a unified query processing pipeline. The query parser identifies both lexical terms and sentiment analysis requests, the retriever fetches documents based on lexical terms, and the sentiment analysis component processes these documents to return sentiment-aware results. This integration allows users to obtain sentiment information directly from search queries without sacrificing search speed.
Solution Approach 2:
The system performs preliminary action by pre-processing queries to identify sentiment analysis components before document retrieval. The query parser预先 identifies which parts of the query require sentiment analysis, allowing the system to efficiently process only the relevant portions of documents rather than analyzing all documents uniformly, thus maintaining speed while providing sentiment capabilities.
3Loss of information
If users read through all hits to extract meaning and insight, then complete information can be obtained, but the process is tedious and inefficient
Solution Approach 1:
The system extracts only the essential sentiment information from documents using automated sentiment analysis algorithms. Instead of requiring users to read through complete documents to extract insights, the system identifies and extracts key sentiment-related information, such as sentiment polarity, intensity, and associated entities, presenting this condensed information directly in the search results.
Solution Approach 2:
The sentiment analysis component acts as an intermediary between the raw document content and the user's information needs. It processes the full document text, identifies sentiment patterns, and transforms this information into structured sentiment-aware results that directly address user queries, eliminating the need for users to manually process entire documents.
4Measurement precision
If sentiment analysis is performed on all retrieved documents, then comprehensive sentiment insights are obtained, but processing time and computational resources increase
Solution Approach 1:
The system applies local quality by performing sentiment analysis selectively on specific portions of documents rather than uniformly on all documents. The query parser identifies sentiment analysis requirements and the system processes only the relevant document sections that pertain to the query's sentiment interests, reducing computational resources while maintaining comprehensive sentiment insights for the actual information needs.
Solution Approach 2:
The system uses partial action by performing sentiment analysis on a subset of retrieved documents rather than all documents. The query parser identifies which documents are most relevant to the sentiment analysis request, and the system processes only those documents, avoiding unnecessary computational resources on irrelevant documents while still providing comprehensive insights for the actual query needs.
Data Source
AI summary
A method, system, and computer program product for information retrieval and sentiment assessment. The method parses a sentiment-aware query to isolate one or more lexical terms to be included in a lexical retrieval of documents containing the lexical terms. The parsing of the query includes parsing the query to isolate portions of the query to be included in the configuration of a sentiment analysis of the retrieved documents. The documents retrieved based on the lexical terms are processed so as to generate a sentiment assessment, and the sentiment found in the retrieved documents might be correlated to terms that are not present in the lexical terms. The sentiment assessment is presented as a “positive” or “negative” indication, or as a sentiment assessment score. The sentiment portion of the query can specify an area of interest, and/or can specify a user-selected classifier model that is used to process the retrieved documents.