KNN Search Algorithm Comparison – This project compares the performance of different K-Nearest Neighbors (KNN) search algorithms across various dataset sizes and dimensions.

A personal project to solve similar problems using various algorithms for performance comparison. For path finding and search problems, implemented algorithms like BFS, DFS, A* and Dijkstra. For various classification and clustering problems, implemented algorithms like KNN, K-Means++, Logistic Regression, LCS, LIS.

Visual comparison of KNN search algorithms on KD trees

Implementation, evaluation, and comparison of recommendation algorithms (SVD, KNN, NMF) using the Surprise library. Includes cross-validation, grid search for model selection, and generation of Top-N personalized recommendations.

—In this paper, we implemented a real-time image recommendation system using ResNet and ANNOY. The aim of this project is to increase the efficiency of the traditional recommendation by decreasing the prediction time in a real-time environment drastically and also make a diverse output that can adapt to the changes int the interest of the users. Our challenge was to make the system work well for the large dataset in real-time and also adjust to new products.Speed is essential in an e-commerce business.Previous methods work using the knn algorithm which is quite slow thus We’ll be using transfer learning, approximate closest neighbors, and the built-in centroid detection in PyTorch to create our recommender.We break down four steps needed to construct the model: Dataset Distribution ,Preprocessing using the normalise and image embeddings , Model we used is the ResNet18 from Transfer Learning and FastAi hooks to retrieve them and at last for search we used ANNOY with the help of Centroid Detection.It help in find the nearest neighbors without pairwise comparison with each dataset thus it is most efficient fit to our model . When customers search the products on the ecommerce website, they aren’t able to get too many similar or equivalent products. So, the customers can miss the many nice clothes and it causes the loss of company also.