India being a fast-developing economy relies heavily on coastal infrastructure to strengthen the overseas trade potential. With mainland coastline of around 6100 km, the construction industry is well placed to support economic growth by developing and upgrading the coastal infrastructure. Coastal facilities include jetties, container yard, railway infrastructure, POL storage terminal, etc. Such facilities impose loads which often are more than the safe bearing capacity of the natural soil. Generally, the soil along the West Coast in the state of Gujarat consists of large saline marshlands, estuaries, and coastal sand dunes from tertiary formations and quaternary deposits. The silty/clayey layer along the West Coast is susceptible to large settlements under structure-imposed loading which could result in failure of infrastructures. In this study, the laboratory test results from multiple sites along the West Coast in the state of Gujarat and Maharashtra are compiled and analyzed to identify patterns using machine learning. The laboratory test results for soft soils are selected and analyzed in machine learning framework for prediction of Cc (compression index) based on clay content, Atterberg’s limits and void ratio. The objective is to predict the compressibility characteristic of soft soils for preliminary design purposes. The prediction philosophy presented in current study will pave a way for initial design check and to recommend ground improvement, if required.