1. Concrete Strength (March 2022 ~ April 2022)

In this project, the regression model is built to predict the concrete strength with given infomration. To do so, the concrete dataset which contains information of Cement, blast furnace slag, fly ash, water, superplasticizer, coarse aggregate, fine aggregate, age, and concrete compressive strength are analyzed. Besides, a few data have a problem that the data is not stated, has a negative value, and stated as NAN. Therefore, the data is cleaned and repair. When fixing the data, the wrong data (empty, negative value, and NAN) is changed to NAN and dropped the columns that the data is stated as NAN. After the dataset is fixed, they are sepearated to train and test dataset to predict the concrete strenght.

For model development, regression models in machine learning schemes are tested and the best model is chosen among them. They are the 3rd degree polynomial linear regression, ANN regression, SVR, Random Forest regression, and ADABOOST regression.


2. Water Potability (April 2022 ~ May 2022)

As the food science technology is developed, many researchers are analyzing elements of the food. The water is one of the most important factors for mankind to sustain their life and keep their body healthy. Therefore, they are trying to classify the water whether it is potable or not. For the drinkable water, it can be determined depending on the ph., hardness, solids, chloramines, sulfate, conductivity, organic carbon, trihalomethanes, and turbidity.

To classify the potability of water with high accuracy, the machine learning methods are used which are Artificial Neural Network (ANN), Random Forest Classifier (RFC), Support Vector Classifier (SVC), AdaBoost Classifier, and Decision Tree Classifier.