Data Mining projects from the Fall 2022 semester
In this assignment, we used the IMDB-Movie-Data dataset and focused on Preprocessing and Normalizing the data using popular libraries such as pandas and scikit-learn. Afterwards, we worked with the GOOGL dataset and made predictions of Google's stock price based on the current day using Linear Regression and Polynomial Regression
In this assignment iris dataset was used and it was mainly about preprocessing and normalizing the data using famous libraries such as pandas and scikit-learn. After that PCA algorithm is used to reduce the dimmension of the data.
This assignment was mainly about working with Tensorflow and Keras for creating different neural networks models to classify make circles and fashion mnist datasets. The impact of not having activation functions or having linear activation functions were also observed. Also, Learning Rate Decay was implemented and the difference between the Adam and SGD optimizer was observed.
This assignment primarily focuses on exploring the k-means clustering algorithm and its implementation for reducing the size of an image. Additionally, we also experimented with the DBSCAN algorithm, and performed parameter tuning to achieve better results.
In this task, we utilized the Apriori algorithm to extract Association Rules from the Hypermarket_dataset dataset. The Apriori algorithm is a popular algorithm for mining frequent itemsets and generating association rules from transactional datasets. It works by iteratively identifying the frequent itemsets in the dataset and then generating association rules based on those frequent itemsets. By applying this algorithm to the Hypermarket_dataset, we were able to extract useful association rules that can reveal the relationships between the purchased items in the dataset.