The principal question for this thesis is as follows: What are the key hydraulic fracturing parameters that affect the amount of hydrocarbons that can be recovered in Eagle Ford Shale? The problem considered in this thesis is establishing a relationship between hydraulic fracturing parameters and the amount of hydrocarbons produced. The importance of this problem is identifying the stimulation parameters that can be used to improve the efficiency and effectiveness of fractures and ultimately increase the production of hydrocarbons. This will assist engineers to make better decisions related to hydraulic fracturing by focusing on the set of key stimulation parameters that are identified in this thesis. The proposed method to the problem is to consider all the stimulation parameters together and use data mining and statistical techniques. In this thesis, the use of four different data mining and statistical approaches, Logistic Regression, Decision Trees, Support Vector Machines, and Neural Networks, are proposed. The foundation is based on the fact that the stimulation parameters are highly interrelated and need be considered together as a whole. These approaches allow the analysis of such a system and have the capability of capturing nonlinear relationships between the input and output parameters. The major findings are identifying eight hydraulic fracturing parameters, Perforated Length Interval (ft), Injection Rate per Stage (bpm), Number of Clusters per Stage, Volume of Proppant per Stage (lbs), Volume of Water per Stage (gals), Number of Stages, Average Treating Pressure per Stage (psi), Maximum Treating Pressure per Stage (psi) as the key stimulation factors that have a direct impact on the amount of hydrocarbons produced, determining an efficient method of analyzing and comparing multiple variables for multiple wells, establishing a production metric that reflects long term production performance, and identifying the best performing regression method.