Semantic Keyword Expansion for Digital Content Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems for identifying instances of digital content are limited by the need for manual specification of search terms, which may not match the metadata keywords of available digital content, resulting in incomplete search results.
Innovation Solution
A computing device implements a content system that receives input data describing attributes of an entity segment and associated keywords, uses a machine-learning model to determine semantically similar keywords, and compiles a set of matchable keywords to identify candidate instances of digital content based on semantic overlap.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual search terms are used to search for digital content, then the search process is simple and direct, but the search results are incomplete when search terms do not match metadata keywords exactly
Solution Approach 1:
A machine learning model is introduced as an intermediary between the search term input and the content matching process. The model automatically generates semantically similar keywords from the input search term, which then serve as intermediaries to match with metadata keywords of digital content, thereby improving search completeness without requiring users to manually specify multiple search terms
Solution Approach 2:
The system changes the parameter of keyword matching from exact string matching to semantic similarity matching. By using a machine learning model to generate and compare semantic representations of keywords, the system can identify semantically equivalent terms even when the exact wording differs, thus improving search result completeness
2Reliability
If exact keyword matching is used for digital content search, then the search process is fast and efficient, but the search results are incomplete when search terms differ in wording from metadata keywords
Solution Approach 1:
The machine learning model performs preliminary action by automatically generating semantically similar keywords before the actual content matching occurs. This preliminary keyword expansion ensures that semantically equivalent content is included in the search results without requiring users to spend time manually specifying multiple alternative search terms
Solution Approach 2:
The system replaces the mechanical process of manual keyword specification and exact string matching with an automated machine learning-based semantic similarity system. The ML model automatically understands and processes semantic relationships between search terms and metadata keywords, eliminating the need for users to manually account for different wording variations
Data Source
AI summary
In implementations of systems for identifying instances of digital content, a computing device implements a content system to receive input data describing attributes of an entity segment and keywords that are associated with the attributes of the entity segment. The content system determines additional keywords that are semantically similar to the keywords using a machine-learning model trained on training data to classify semantically similar keywords. A set of matchable keywords is compiled that includes the keywords and the additional keywords. The content system identifies candidate instances of digital content based on content keywords assigned to the candidate instances of digital content and the set of matchable keywords. An indication of an instance of digital content is generated for display in a user interface based on the candidate instances of digital content.