Pinned Repositories
SpeakingFaces
A large-scale publicly-available visual-thermal-audio dataset designed to encourage research in the general areas of user authentication, facial recognition, speech recognition, and human-computer interaction.
Airbus_ship_segmentation
A-Study-of-Sigmoid-based-Multi-class-Logistic-Regression
Investigated how multi-class logistic regression would perform if the activation function was changed from softmax to sigmoid. It included mathematical analysis and empirical evaluation, such as rewriting the model from scratch. Tech: Python (Scikit-Learn, Pandas)
Customer-Clustering
Conducted customer clustering, made detailed description for each customer group, and made churn analysis of the customers.
Extract-Pop-Up-Window-Information-from-MMA-Fights
Given a video of an MMA fight, extract information from all pop-up windows that appear anywhere in the video.
Investigating-Perceptron-Learning-Algorithms
Rewrote PLA and Pocket PLA algorithms from scratch, and analyzed their performance against Linear Regression model. Tech: Python (Pandas, NumPy)
Predicting-Life-Satisfaction-Rate-with-Multiple-Linear-Regression
Analyzed what factors influence the level of satisfaction of people living in developing countries. The analysis contained predictor subset selection with VIF, Mallow’s Cp and BIC scores, model validation, and remedial measures. The model performance was evaluated based on hypothesis testing. Tech: R (RStudio)
Red-Wine-Quality-Classification-using-Supervised-Learning
Conducted data analysis using statistical tools and complex visualizations; trained logistic regression, k-nn, kernelized svm, and random forest models; performed hyperparameter tuning and error analysis. Tech: Python (Seaborn, Matplotlib, Pandas, Scikit-Learn)
SaniyaAbushakimova's Repositories
SaniyaAbushakimova/Predicting-Life-Satisfaction-Rate-with-Multiple-Linear-Regression
Analyzed what factors influence the level of satisfaction of people living in developing countries. The analysis contained predictor subset selection with VIF, Mallow’s Cp and BIC scores, model validation, and remedial measures. The model performance was evaluated based on hypothesis testing. Tech: R (RStudio)
SaniyaAbushakimova/A-Study-of-Sigmoid-based-Multi-class-Logistic-Regression
Investigated how multi-class logistic regression would perform if the activation function was changed from softmax to sigmoid. It included mathematical analysis and empirical evaluation, such as rewriting the model from scratch. Tech: Python (Scikit-Learn, Pandas)
SaniyaAbushakimova/Investigating-Perceptron-Learning-Algorithms
Rewrote PLA and Pocket PLA algorithms from scratch, and analyzed their performance against Linear Regression model. Tech: Python (Pandas, NumPy)
SaniyaAbushakimova/Red-Wine-Quality-Classification-using-Supervised-Learning
Conducted data analysis using statistical tools and complex visualizations; trained logistic regression, k-nn, kernelized svm, and random forest models; performed hyperparameter tuning and error analysis. Tech: Python (Seaborn, Matplotlib, Pandas, Scikit-Learn)
SaniyaAbushakimova/Customer-Clustering
Conducted customer clustering, made detailed description for each customer group, and made churn analysis of the customers.
SaniyaAbushakimova/Extract-Pop-Up-Window-Information-from-MMA-Fights
Given a video of an MMA fight, extract information from all pop-up windows that appear anywhere in the video.