Phrase Association Search Ranking on Social Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social networking systems face challenges in presenting relevant search results that accurately match user queries, as existing methods struggle to identify and rank posts based on the intended topic, leading to irrelevant or low-quality search outcomes.
Innovation Solution
The system analyzes user queries by identifying candidate phrases, filtering out posts without known phrases, and ranking remaining posts based on the degree of equivalence between query terms and known phrases, using features such as Normalized Pointwise Mutual Information (NPMI) scores and aggregation models to determine the relevance of search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the system retrieves posts containing query terms using existing search methods, then search coverage is improved, but search result relevance deteriorates
Solution Approach 1:
The system performs preliminary phrase association analysis offline to build a phrase association table before search execution. This pre-computed statistical information (NPMI scores) is stored and reused during online search, enabling fast relevance ranking without real-time computation overhead, thus maintaining both broad coverage and high relevance
Solution Approach 2:
The patent introduces a phrase association table as an intermediary data structure between the query and post retrieval. This table stores pre-computed NPMI scores for phrase pairs, acting as a mediator that translates query terms into relevance-ranked results without requiring complex real-time analysis, thereby improving relevance while maintaining efficiency
2Measurement precision
If the system performs detailed phrase analysis to improve search accuracy, then search result relevance is improved, but computational complexity increases
Solution Approach 1:
The system performs the computationally intensive phrase association analysis (NPMI calculation) offline before search execution. The pre-computed phrase association table is stored and reused for multiple searches, eliminating the need for real-time computation during user queries, thus achieving high accuracy without increasing online computational complexity
Solution Approach 2:
The system uses its own social graph data and post content to automatically build and maintain the phrase association table without external intervention. The statistical analysis is performed on existing data within the system, making the complex computation self-contained and reusable across multiple search operations
Data Source
AI summary
In one embodiment, a method includes retrieving multiple posts containing each of the query terms after receiving a search query for posts from a user of an online social network, identifying, from an index of known phrases, one or more known phrases in the text of the retrieved posts, filtering the retrieved posts to generate a filtered set of posts, wherein the filtering comprises removing each post not containing at least one of the identified known phrases, computing features for each of the identified known phrases, calculating a score for each post in the filtered set based on the features of the identified known phrases, ranking the posts based on the calculated scores, and sending instructions, to the client system, for generating a search-results interface comprising references to one or more of the posts presented in ranked order.