Name : Yashi Srivastava
ID Number : 2000123
Project Description: The project will aim on creating a predictive model on one of the dataset present in public domain. Initially, I intend to start with amcat dataset, which has scores of candidates in various domains, and their past academic profile, and placement package is given. The predictive model should be capable of predictive the placement package range on new set of entries with a good accuracy.
Tasks Associated:
- Data - Preprocessing
- Feature Reduction
- Feature Scaling
- Testing various Classification algorithm
Challenges:
- Over 40 features
- Only 4000 tuples, which may be insufficient to provide insight on such large feature count
Work Plan: Will build a basic model on the data first, without getting a lot into the first 3 tasks. If the algorithm will seem to give good results, will continue on this work, with a more emphasis on detailing, otherwise will plan something else accordingly.