Media Recommendation via Mixed Word Sequence Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional media content recommendation methods fail to accurately match user interests due to the lack of consideration for new vocabularies, resulting in low recommendation accuracy.
Innovation Solution
A media content recommendation method and apparatus that performs word segmentation, combines adjacent words to generate a mixed word sequence, allocates matching weight coefficients to phrases, and searches for media content, considering both the relevance of segmented words and the mixed words to improve recommendation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only simple word segmentation is used for matching reference, then the system is simple and easy to implement, but recommendation accuracy is low because new network vocabularies cannot be matched
Solution Approach 1:
The patent applies segmentation by dividing the input entry parameter into multiple word sequences using different segmentation granularities. The first word sequence uses fine-grained segmentation while the second uses coarse-grained segmentation, allowing the system to capture both individual words and multi-word phrases including new network vocabularies, thereby improving recommendation accuracy without excessive complexity
Solution Approach 2:
The patent creates a composite matching reference by combining multiple word sequences with different segmentation levels. This composite structure integrates fine-grained and coarse-grained segmentation results, enabling the system to match both individual words and multi-word phrases simultaneously, thus resolving the contradiction between simplicity and accuracy
2Measurement precision
If multi-word phrases and new vocabularies are considered in matching, then recommendation accuracy improves, but the complexity of word processing and matching increases
Solution Approach 1:
The patent segments the input parameter into multiple word sequences with different granularity levels. The first sequence captures individual words while the second sequence captures multi-word phrases, allowing the system to process both fine-grained and coarse-grained elements systematically, improving matching accuracy for new vocabularies while maintaining manageable processing complexity
Solution Approach 2:
The patent creates a universal matching reference that serves multiple functions: it matches individual words, multi-word phrases, and new network vocabularies simultaneously. This multi-functional approach eliminates the need for separate processing pipelines for different phrase types, improving matching comprehensiveness without proportionally increasing processing complexity
Data Source
AI summary
A media content recommendation method includes: receiving an input entry parameter, and performing word segmentation on the entry parameter to obtain a word sequence; combining adjacent words in the word sequence to generate a mixed word sequence; acquiring phrases included in the word sequence and the mixed word sequence, and separately allocating a corresponding matching weight coefficient to a phrase; searching for media content including the phrase, acquiring a matching weight coefficient sum of phrases included in the found media content, and screening the found media content according to matching weight coefficient sums and performing recommendation. In addition, a media content recommendation apparatus is also provided.