Kaggle Solution Review: Rate Text Readability

Meeting agenda: (1) Update Exercise (2) Book Club

UPDATE EXERCISE Begin with sharing your usual check-in sheet (what is the best, worst thing professionally, personally last month, next month) (8 bullet points)

KAGGLE SOLUTION REVIEW Kaggle, a subsidiary of Google LLC, is an online community of data scientists and machine learning practitioners. Kaggle allows users to find and publish data sets, explore and build models in a web-based data-science environment, work with other data scientists and machine learning engineers, and enter competitions to solve data science challenges. We are doing Kaggle Solution Review in order to become familiar with the state of the art models in Data Science. During this meetup we will look at CommonLit Readability competition, which asked participants to identify the appropriate reading level of a passage of text. The target value is the result of a Bradley-Terry analysis of more than 111,000 pairwise comparisons between excerpts. Teachers spanning grades 3-12 (a majority teaching between grades 6-10) served as the raters for these comparisons. In this competition, data scientists built algorithms to rate the complexity of reading passages. We will look at the dataset, discuss traditional methods for deducing text complexity based on such proxies as characters or syllables per word or number of words per sentence, and then look at the proposed solutions to the competition.

Important links: Competition description: https://www.kaggle.com/competitions/commonlitreadabilityprize/overview Data Sources: https://www.kaggle.com/c/commonlitreadabilityprize/discussion/240423 1st place solution: https://www.kaggle.com/competitions/commonlitreadabilityprize/discussion/257844