Sound velocity detection at sea using machine learning techniques. (2021)

• Record Type: Journal Article
• Authors: Anirudh, Rani Venkata Satya; Pattanaik, Bishal; Lakshmeeswari, G.; Vamsy, G.
• Is Part Of: Materials today. Volume 45:Part 2(2021)
• Page Start: 2953
• Page End: 2957
• Publication Date: 2021
• Subjects: Multi beam echo-sounder -- Data mining -- Linear regression -- Random forest -- Stochastic gradient descent; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2020.11.948
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: In this research paper, various machine learning techniques are implemented to predict the velocity of sound at sea. An analysis of various potential predictors and their correlations are also underlined. The dataset used for this research is a synthetic dataset. Multi beam echo sounder, an instrument that calculates various parameters like temperature, salinity, depth, etc. is used for making the dataset. This equipment is generally used on research ships to study the seawater characteristics in ocean at various places. It was permitted for use from such an Indian Coastal Research Vessel (CRV) on request for academic purpose only. The methodology proposed in this paper learns the pattern of every observation using the following algorithms: Linear Regression, Random Forest and Stochastic Gradient Descent (SGD). Furthermore, an analysis of the features and their correlations with the target variable, with the help of various Data Mining methods is presented.