There are a lot of pattern recognition techniques which help us analyze data and come up with valuable, re- fined information and predictions. The most common used techniques include Principle Component Analysis(PCA), Linear Discriminant Analysis(LDA), Naive Bayes, K-Nearest Neighbors(KNN), and Support Vector Machine(SVM) variants. These techniques help us bringing out hidden patterns and makes it easier for us to interpret our data.
Experiments have been conducted with PCA, LDA, Naive Bayes, KNN, SVMs and Dense Neural Networks on House Price data-set in order to better represent the effects of each method on a real world, time-series data with both categorical and numerical values as features.
It has been concluded that a dense neural network with five hidden layers yields the best result with 85% among all methods mentioned.
- numpy == 1.18.*
- pandas == 1.0.*
- matplotlib == 3.2.*
- tensorflow >= 2.0.0
- sklearn == 0.22.*
- seaborn == 0.10.*
- Python == 3.6.*
Clone the github repository.
git clone https://github.com/NickJackolson/HousingPricesML.git
pip3 install -r requirements.txt