This repository is used to calculate two major metrics of the Recommender System UI testing app:
- Task Completion Time (TCT)
- Click-Through-Rate (CTR)
- recommender-system-ui: Main Amazon-like React Native testing app
- recommender-system-scraper: Scripts to obtain data from Amazon
average_ctrs_df:
interfaceId ctr
0 control 0.000000
1 large_long 0.058463
2 large_short -0.010000
3 small_long -0.010952
4 small_short 0.020168
average_tcts_df:
interfaceId tct
0 control 0.0000
1 large_long 29.2979
2 large_short 8.6066
3 small_long 2.3894
4 small_short 14.2318
average ctr: 0.21142857142857138
average tct: 51.4721
To cite this project, please consider using the following bibtex:
@misc{recommender-system-analysis,
author = {San Juan, Justin and Chambers, Owen},
title = {Tailoring Product Recommendations for Mobile Devices},
year = {2022},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/justinsj/recommender-system-analysis}},
}