Changes between Version 6 and Version 7 of AdvancedTopics
- Timestamp:
- 06/15/26 02:37:24 (27 hours ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
AdvancedTopics
v6 v7 1 = Recommendation System for worker feed with vector database = 1 For the advanced topic, we used a vector database for two use cases in our application: (1) to provide a personalized feed for workers by recommending task requests similar to those they have previously accepted and completed, and (2) to recommend suitable workers to clients when they create a new task request. 2 = 1. Recommendation System for worker feed with vector database = 2 3 3 4 For our recommendation system, we used a vector database so we can recommend task requests on the worker's feed so that they are closest to the tasks that the worker usually takes on. … … 7 8 We will look at how both of the strategies are implemented. 8 9 9 == 1. Personalized Recommendations ==10 == 1.1 Personalized Recommendations == 10 11 11 12 This is the recommendation system that recommends open task requests to workers that have the most similar completed tasks. For this purpose we first embedded the open task requests and then as a starting point for our system we embedded three task requests that resulted in a completed task for each worker. We found this number of tasks enough as a starting point for our app. We built these embeddings with the script [attachment:embed_taskrequests.py] and we used a table to store these vectors: … … 274 275 For this approach to work, every new task request submitted by a client must be embedded before it can participate in the recommendation process. 275 276 276 == 2.Category Based Recommendation ==277 == 1.2 Category Based Recommendation == 277 278 278 279 If the worker has no completed tasks we will use the experience description that the worker provided when signing up for a category. … … 317 318 }}} 318 319 319 = Worker Recommendation for Task Requests =320 = 2. Worker Recommendation for Task Requests = 320 321 321 322 Besides recommending task requests to workers, the system also supports recommending suitable workers for a specific task request.
