1 / 8

Enhancing eCTR Prediction with Latent Topic Models and User Clustering Techniques

This project outlines a high-level architecture for improving estimated click-through rates (eCTR) using latent topic modeling, user encoding, and clustering frameworks. The pipeline includes encoding user data through an auto-encoder for dimension reduction and clustering users based on political affiliations. The output will be stored in a Hive table containing user IDs and their low-dimensional representations. Additionally, optional user clustering stages will be explored to refine prediction accuracy. Proposed models and algorithms are detailed alongside project plans and resource allocation.

trista
Download Presentation

Enhancing eCTR Prediction with Latent Topic Models and User Clustering Techniques

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Plans on “Latent Topic Model”

  2. High-Level Architecture Users Ads User Encoding UserEncoding UserClustering Prediction eCTR / FB Prediction

  3. Existing Pipeline • Encoding • Auto-encoder for dimension reduction • Political affiliation clustering • Output: Hive table (user id + low-dim representation) • eCTR prediction • Optional: user clustering stage

  4. Approaches to use encoding in eCTR prediction

  5. Proposed Models

  6. Algorithm Details

  7. Project Plans

  8. Resources

More Related