Academic Projects

Dive into the intriguing world of Expert Systems (ES) with this hands-on project focused on healthcare. Discover the essentials of ES, including rule-based reasoning and key components, using Pyke, a powerful Python-based knowledge engine. Learn how to leverage logic programming in the development of an ES tailored for healthcare applications. Join us as we delve into the potential, limitations, and areas for improvement of ES across healthcare and other real-world sectors.

Journey through heuristic-based problem solving using the A* algorithm and SimpleAI. Solve the classic 8-Puzzle problem, delve into the power of heuristics, explore the Manhattan distance heuristic’s efficiency, and discuss real-world applications.

Explore the use of Naive Bayes and scikit-learn in predicting golf play based on weather. Delve into Bayes' theorem, Laplacian correction, and the practical application of the categoricalNB() model. Uncover the significance of frequency tables, likelihood tables, and posterior probabilities in making accurate predictions.

Discover the development of a 2-layer Artificial Neural Network (ANN) in Python, designed to predict the next number in a user-input series. Learn about key elements of ANNs, such as backpropagation, forward-propagation, and loss calculation. Examine the project’s performance on various data sequences and explore potential improvements for enhanced functionality.

Explore the world of k-Nearest Neighbors classifiers (k-NN-C) with this comprehensive project, where we build a high-performing model using the Iris Dataset and Python. Learn about the inner workings of k-NN-C, lazy learning, and the importance of Euclidean distance in generating predictions. Dive into the process of constructing the model, including preprocessing the dataset and using cross-validation splits. Achieve an impressive mean accuracy of 96.67% with 5-fold cross-validation and discover the model’s potential applications in various classification tasks, along with future research recommendations.

Dive into the world of Simple Linear Regression using Scikit-Learn and Python to predict the median value of owner-occupied homes in the Boston housing dataset. Explore the dataset and its features, perform Exploratory Data Analysis (EDA), transform variables to prevent underfitting, implement Simple Linear Regression, and analyze the predictions. Understand the impact of various factors on housing prices, and learn how to create a powerful prediction model using Python and Scikit-Learn.

Delve into the world of balanced binary search trees, specifically AVL trees, and learn how to implement them using Python. Understand the fundamental differences between binary search trees and AVL trees, as well as their benefits in terms of search, insertion, and deletion operations. Examine the Node and Tree classes, the build_tree() method, and the insert() and delete() methods in detail. Explore the program implementation, testing process, and performance analysis of AVL trees in comparison to traditional binary search trees. Discover the applications of AVL trees in various data manipulation tasks and real-world scenarios.

Explore the fascinating world of word embeddings using Word2Vec, TensorFlow 2.0+, and Python to build a cutting-edge model from Wikipedia data. Dive into the intricacies of word representations, word embeddings, and the word2vec software. Examine the TensorFlow demo, its model, and dataset, while learning about the unsupervised skip-gram negative-sampling model. Understand the potential applications in natural language processing tasks such as chatbot development and learn how to create powerful word embeddings using TensorFlow and Python.

Discover the power of text dataset augmentation using round-trip translation, Python, and the NLPAug library. Learn about various text augmentation techniques, such as synonym augmentation, semantic similarity augmentation, and round-trip translation. Explore the capabilities of the NLPAug library, and see how a Python script can utilize it to augment text datasets for natural language processing tasks like chatbot training. Dive into the script’s implementation and understand the potential applications of text augmentation in improving language models.

Explore the potential of feature engineering and hyperparameter tuning in improving text classification models for identifying abusive content online. Delve into feature extraction, selection, construction, and the bag-of-words (BOW) approach. Examine character- and word-level representations, term frequency-inverse document frequency (TF-IDF), tokenization, stopwords, and stemming. Learn about Support Vector Machines (SVMs), their hyperparameters, and tuning techniques, such as grid and random searches. Understand the role of feature engineering in boosting the performance of hate speech detection algorithms and SVMs.