Emoji Recommendation via Bayesian Category Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing emoji input methods suffer from low efficiency due to manual selection from large emoji lists and low probability of emoji recommendation triggers, leading to frequent reliance on manual selection.
Innovation Solution
An emoji recommendation method that clusters training sentences to create interpretive categories, calculates probabilities using the Bayes formula to determine the target category for user input, and recommends emojis based on these probabilities, improving accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the user manually selects an emoji from a large list, then the user can choose any emoji, but the input efficiency is low
Solution Approach 1:
The system automatically analyzes the user's input text and proactively recommends emojis without requiring manual search or selection from a large list. The emoji recommendation module detects keywords and contexts in real-time, automatically matching them with appropriate emojis from the database, thereby making the system serve the user's needs autonomously while maintaining high versatility in emoji selection
Solution Approach 2:
The system pre-configures a comprehensive emoji database with multiple tags and categories before the user needs to input emojis. By organizing emojis in advance with detailed metadata and classification, the system prepares the foundation for rapid automatic matching during actual use, eliminating the need for manual browsing while ensuring all possible emoji options are available
2Ease of operation
If the system recommends emojis based on tag matching, then the recommendation process is simplified, but the triggering probability is low
Solution Approach 1:
The system expands the matching parameters beyond simple tag equality to include keyword matching, contextual analysis, and semantic similarity. By changing the matching criteria from exact tag matches to multi-parameter evaluation (including word frequency, position, and context), the system significantly increases the probability of triggering recommendations while maintaining ease of operation
Solution Approach 2:
The recommendation system dynamically adjusts its sensitivity and matching thresholds based on the user's input context. Instead of using fixed tag-matching rules, the system adapts its detection parameters in real-time according to the length, type, and content of the user's text, thereby increasing the triggering probability across diverse input scenarios while keeping the interface simple
3Device complexity
If the system uses simple tag matching for emoji recommendation, then the system complexity is reduced, but the recommendation accuracy is low
Solution Approach 1:
The system segments the emoji database into multiple categories and sub-categories with hierarchical tag structures. By dividing the large emoji set into manageable segments (e.g., emotions, objects, actions, contexts), the system can apply targeted matching rules to each segment, improving accuracy without proportionally increasing overall system complexity
Solution Approach 2:
The system introduces an intermediate keyword extraction and weighting layer between the user's text input and the emoji database. This intermediary module processes the input text to identify key semantic elements and their importance weights, then uses these as mediators to guide the emoji matching process, thereby improving recommendation accuracy while keeping the core system architecture relatively simple
Data Source
AI summary
The present disclosure provides an emoji recommendation method and device. The emoji recommendation method may include: acquiring training sentences corresponding to emojis; clustering the training sentences to categories; according to a number of training sentences in each category, calculating a first probability corresponding to each category; according to correspondences between the training sentences and the emojis, determining emojis included in each category, and calculating a second probability corresponding to each emoji in each category; according to occurrences of each word in each category, calculating a third probability corresponding to each word in each category; according to the first probability and the third probability, calculating a fourth probability of a target sentence belonging to each category; according to the fourth probabilities corresponding to the categories, determining a target category for the target sentence; and according to the second probability corresponding to each emoji in the target category, recommending an emoji.