International Journal of Computer Science Trends and Technology (IJCST) – Volume 5 Issue 3, May – Jun 2017 ISSN: 2347-8578 www.ijcstjournal.org Page 17 Regression And Augmentation Analytics on Earth’s Surface Temperature Siddharth Banga [1] , Saksham Mongia [2] ,VaibhavTiwari [3] Mrs. Sunita Dhotre [4] Student [1], [2] & [3] , Associate Professor [4] Department of Computer Engineering Bharati Vidyapeeth University College Of Engineering, Pune Maharashtra, India I. INTRODUCTION Earth’s temperature observation plays an important role in Big Data Analysis. This provides useful information of Climatic change through which detection of the future problem like Global Warming, melting of Glaciers and many more can be overcome for the sake of future generations. This study is based upon the large data sets starting from year 1901 to 2015.Evaluation of monthly temperature of every year and then plotting the graph of temperature change that has happened after every 10 years using Linear Regression [8]. This Earth’s temperature program will help the future Earth’s program with an important role in academic’s and decision making for sustainable environment. Table 1 (Values in Degree Celsius) II. METHODS/APPROACH Temperature Anomaly with regression analysis Temperature anomaly can be defined as a divergence from a reference value or long-term average [6]. It can be classified into positive and negative anomalies where from a positive anomaly demonstrates that temperature was warmer than the reference value, while a negative anomaly demonstrates that the observed temperature was cooler than the reference value [10]. It is a diagnostic tool for global scale climate which provides a big picture overview of average global temperature with a reference value [7] . Regression analysis: regression analysis is a statistical process for estimating the relationships among variables [5] . Regression line formula x on y-axis is defined as: ………. (1) Regression line formula x on y-axis is defined as: ………. (2) Estimation of global average temperature is difficult that is why elevation of temperature of region is also considered. For its representation over years process of Normalization and computation of reference values a base line is established on which temperature anomalies is processed [13] . In Table 1, average land temperature and average land uncertainty of time span of 10 years is calculated for over a period of 100 years i.e. 1901-2001. ABSTRACT Big Data Analysis is the process of handling the huge amount of data that overcomes hidden patterns, market trends and other useful information that can help organization to make better decision [2] . GISS temperature scheme was defined in 1970’s by James Hansen when method of global temperature was needed. The scheme was based on finding the co-relation of temperature change between the stations separated by 1200km [9] .These facts were sufficient to obtain useful estimate for global mean temperature change .Our study is necessary to define the co-relation between Land’s Average Temperature and Uncertainty temperature, so as to show the increasing temperature year after year which is the major cause for Global Warming [3] . Keywords:- Uncertainty, Linear Regression, Anomaly, precision. Year *Land Average Temperature *Land Average Uncertainty 01-01-1901 8.224833 0.275606 01-01-1911 8.29442 0.25853 01-01-1921 8.52015 0.249766 01-01-1931 8.65478 0.239741 01-01-1941 8.68547 0.217308 01-01-1951 8.64268 0.15295 01-01-1961 8.64484 0.097191 01-01-1971 8.68634 0.097775 01-01-1981 8.936875 0.08843 01-01-1991 9.151833 0.078408 01-01-2001 9.56482 0.085483 RESEARCH ARTICLE OPEN ACCESS