Feature selection using graph-based clustering for rice disease prediction
Learn more
Created an end-to-end project from taking sensor data to storing data temporarily in a Kafka server to pushing the data into MongoDB and finally deploying in AWS. This is a Binary Classification problem, in which the positive class indicates that the failure was caused by a certain component of the Air Pressure System of a truck, while the negative class indicates that the failure was caused by something else.
Learn moreThis package is meant to quickly preprocess text data to save developers time. This package includes lemmatization, stemming, punctuation removal, stopwords removal and many more.
Learn moreDeveloped artificial intelligence (AI) that can scan facial photos and determine whether or not a person is wearing a face mask, as well as the sort of mask being worn.
Learn moreCreated an end-to-end project starting from data ingestion, preprocessing, model selection, model tuning, prediction, logging framework, deployment, and model retraining to predict whether a person is suffering from thyroid or not.
Learn moreThis project is designed to recommend the food items and recipes based on the user choice, nutrition needs, and healthy foods.
Learn morePredict cars based on images using transfer learning technique. Here scraping and data augmentation is also used to get more images
Learn moreAll my publications done during my undergrad.
Learn more
All my experiences as a Data Scientist/Machine Learning Engineer
• Implemented an enhanced menu similarity analysis using RNN, resulting in a 12 percent accuracy improvement over the previous model.
• Conducted web scraping using Beautiful Soup and Selenium to extract office locations for clients from the Spacelist website across 20 cities in Canada, facilitating efficient business operations and decision-making for clients seeking suitable office spaces.
• Implemented a robust triangle detection strategy, leading to a 15 percent increase in trading gains.
• Utilized advanced forecasting techniques such as ARIMA and SARIMAX models to accurately predict future trends, achieving a prediction accuracy of 73 percent.
• Utilized CNN techniques to forecast stock trends and achieved an accuracy rate of approximately 81 percent through extensive back-testing.
• Gathered raw data from various sources and performed data cleaning operations, resulting in improved data accuracy.
• Led initial data exploration, utilizing visualization and statistical analysis.
• Demonstrated strong proficiency in feature engineering by creating novel and impactful features.
• Implemented robust log creation and custom exception handling for reliable program execution.
• Developed and implemented more than 3000 lines of code for recommendation project, showcasing strong programming skills and attention to detail.
• Successfully performed data scraping from over 100 web pages, demonstrating the ability to gather and handle large-scale data efficiently.
• Conducted extensive data cleaning and preprocessing on a substantial food dataset, consisting of over 2 million records. This process ensured high-quality data for subsequent analysis.
• Exhibited expertise in machine learning by performing rigorous model selection, fine-tuning, and optimization, resulting in a remarkable 18% improvement in model performance.
• Played a pivotal role as a contributing member of the AIOPS (Artificial Intelligence for IT Operations) team, actively contributing to the development and implementation of cutting-edge AI solutions.
• Actively contributed to a research paper titled "A Survey on Rice Plant Disease Identification using Image Processing and Data Mining Techniques, " demonstrating a passion for both agriculture and technology.
• Played a key role in coding and implementing various image processing and data mining algorithms, aiding in the identification and analysis of rice plant diseases.
• Assisted in the extensive data collection effort by capturing and curating a comprehensive dataset of more than 10,000 images encompassing diverse varieties of Rice Plant Diseases.
• Collaborated effectively with the research team, leveraging technical expertise and domain knowledge to enhance the accuracy and robustness of disease identification methodologies.