kunal1406's Projects
Implementation of Computer Vision
Devised a hierarchical clustering framework for feature extraction from scratch, applied to 90K+ rows data using cluster centroids. Transformed data enhanced KNN classifier performance from 90% to 97%, confirmed via K-fold cross-validation & base classification comparison.
Consist of data files
Decision Trees and Ensemble Classifiers - Develop and assess decision trees with varying depths, along with bagging and AdaBoost ensemble methods, for predicting gender using height, weight, and age data in order to analyze performance and overfitting.
Used Gemini Pro for Resume ATS score and keywords to find the content to tailor your resume
Convolution Neural Network plus Computer Vision
Modeling Global and Local Node Contexts for Text Generation from Knowledge Graphs (authors' implementation for the TACL20 paper)
Application of LLM model
Collection of LeetCode questions to ace the coding interview! - Created using [LeetHub](https://github.com/QasimWani/LeetHub)
Implementation of LSTM on a large dataset and anomaly detection
Implementation of Autoencoder using LSTM architecture
Random Forest Hyperparameter tuning using grid search
Implementing and analyzing linear, locally weighted linear, and logistic regression techniques to fit and predict 1D data, trigonometric basis functions, and gender classification.
Exploring Maximum Likelihood Estimation (MLE), Maximum a Posteriori (MAP) for Poisson processes, and implementing K-Nearest Neighbor and Gaussian NaΓ―ve Bayes classifiers for gender prediction.
Send alerts via SMS if any anomaly detected in temperature or if value goes below minimum value, Also, predicts the temperature values.
Explored unsupervised and semi-supervised learning techniques using hierarchical clustering and self-training with K-nearest neighbor classifier on height, weight, age, and gender data.