Identify clusters concerning demographics and retrieve what makes each group happy with HappyDB.
This analysis in R is based on the open dataset HappyDB (https://rit-public.github.io/HappyDB/). It includes the data of 10843 people and 100535 happy moments, which were collected using Amazon Mechanical Turk.
The analysis in R is basically divided into the following steps:
1.a Data Exploration & Text Analysis of the Happy Moment Dataset (credits to: Chen Chen https://www.kaggle.com/powderist/happydb-analysis , who did the Text Analysis in Python)
1.b Data Exploration & Clustering of the Demographic Dataset
2. Correspondence Analysis on the Joined Dataset
3. Classificatin using an Conditional Inference Tree
4. Assessment of the results & future work
Any improvements are highly welcome! Most importantly, have fun! :)
Authors: Patricia Lucía Caruana Blasco, Antoine Lain, Jacqueline Neef