A machine learning based approach to study the fertilization of sea urchin on varying pH
Our objective is to implement a machine learning-based approach to predict the fertilization of sea urchin on varying pH. The dataset has 26 features such as site properties, pH, and sperm concentration. Information regarding the dataset can be found on the site mentioned below. We studied dataset through data visualisation techniques and Exploratory Data Analysis(EDA). We performed PCA analysis for further optimisation. Our target variables are Normally fertilised eggs, Unfertilised eggs, Abnormal fertilised eggs, Abnormal eggs. We applied suitable machine learning models to optimise the predictability of the target variables.
Knowledge of different machine learning implementation models such as Linear Regression, Random Forest, Neural Network, KNN regression. A basic understanding of Sea Urchin, how fertilisation is affected by certain factors such as temperature, sperm concentration etc. Participants should know the intermediate level of python implementation and basic data visualisation methods. It is a great read for nature enthusiasts.
Draft Poster https://drive.google.com/open?id=1g_-qJKr2B_sHXSS4jo_AQIotuPm4R1aj https://github.com/iitgoa-ml/marine-data-science
Dataset from: https://doi.pangaea.de/10.1594/PANGAEA.872634
Abhilasha Gupta is a Ph.D. student in IIT Goa. Her domain is Machine Learning. Raj Hansini and Neeraj Khatri are 3rd-year Computer Science Engineering undergraduates at IIT Goa. Devraj Mogaveera is a 3rd-year Electrical Engineering undergraduate at IIT Goa. This project was supported by the IIT Goa Summer Internship Program 2019 and guided by Dr. Clint P. George, Assistant Professor, IIT Goa.