Aspect Extraction from Text via SVO Parsing and N-gram Frequency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods are inefficient in extracting valuable information from large volumes of unstructured digital text, such as online reviews and forum posts, making it impractical for users to access and utilize the latent opinions and sentiments within.
Innovation Solution
A system and method for extracting aspects from text streams by parsing sentences into subject-verb-object triples, identifying relevant verbs, tokenizing and normalizing subjects, generating n-grams, and selecting aspects based on frequency, to automatically extract and summarize opinions from text data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If manual review of large volumes of unstructured text is performed, then complete information access is achieved, but time consumption and practicality deteriorate significantly
Solution Approach 1:
The patent extracts only the essential aspect information from unstructured text reviews, pulling out key product attributes and opinions while discarding redundant content. This extraction approach enables users to access critical information without reading entire reviews, resolving the contradiction between information completeness and time efficiency.
Solution Approach 2:
The system creates structured summaries that copy and condense the essential meaning of multiple unstructured reviews into standardized aspect-based formats. These summaries preserve the core information while dramatically reducing the time required to process and understand review content.
2Productivity
If automated aspect extraction is implemented, then processing efficiency is improved, but system complexity increases
Solution Approach 1:
The patent segments the text processing task into distinct stages: sentence parsing into SVO triples, verb identification and matching, subject extraction as aspect candidates, tokenization, normalization, and n-gram generation with frequency analysis. This segmentation enables efficient automated processing while managing system complexity through modular design.
3Measurement precision
If frequency-based n-gram selection is used, then aspect extraction accuracy is improved, but computational requirements increase
Solution Approach 1:
The patent transforms aspect candidates through tokenization and normalization to create standardized n-grams, then uses frequency analysis to identify significant aspects. By changing the parameter from raw text to normalized n-grams and using frequency as a selection criterion, the system achieves accurate aspect extraction with manageable computational requirements.
Data Source
AI summary
A text stream having one or more sentences is received, and any number of the one or more sentences are parsed to determine corresponding subject-verb-object (SVO) triples. Each sentence whose corresponding SVO triple contains an identified verb is selected, based on the identified verb, or a lemma of the identified verb, matching a predefined verb. A subject of each selected sentence is identified as an aspect candidate. Each identified aspect candidate is tokenized and normalized. One or more n-grams are generated for each tokenized and normalized aspect candidate. For each generated n-gram, a frequency at which the n-gram is generated is determined. A number of the generated n-grams are selected as aspects based on the frequency with which the number of n-grams are generated.