Trouble404 / kaggle-Job-Salary-Prediction

Advanced Machine Learning group coursework

kaggle-Job-Salary-Prediction

Advanced Machine Learning group coursework

Describtion of Adzuna Salary Prediction here

Aim of this project:

• Using different method to predict
• Comprasion

Step 1: data_cleanning - fixed -> reduce the dimention of input data

fill NaN and transofom strings to integer vector. At the sametime, update location information by mathching LocationTree.

orginal data format:

after step 1

Processed data can be downloaded at link：https://pan.baidu.com/s/1DrLCkmjK4GxdOsHHnDpEZw pin：8nk1

Step 2: Genreate one-hot vector

transfrom vector to one-hot vector and using Embedding or HashingVectorizer to process data in FullDescription

Step 3: Train model

try to use data without fulldescription right now, meet overfitting. Perpare to pre-processing salay data first.