Shilpa Mangesh Pande et al., International Journal of Emerging Trends in Engineering Research, 8(7), July 2020, 3236 - 3240 3236 ABSTRACT Agriculture is one of the fundamental occupations for majority of the countries in the world. Especially, in developing nations like India, the country is primarily driven by agriculture sector, where agriculture and its associated businesses are the backbone of the Economy making it the integral revenue generator. With technological advancements in the recent years, crop yield prediction has gained wide importance, and has shown to have significant impact on the revenue generated from agriculture in every season. Multiple factors influence crop yield prediction, which in turn makes it a non-trivial and challenging task. Despite many proposed works in the area, crop yield prediction lacks a unified solution. This paper brings out the need for a unified framework through a comparative study of standard algorithms and attributes. The algorithms considered are Linear Regression, Random Forest, K Nearest Neighbors (KNN) and Stochastic Gradient Descent (SGD). Our results show that Random Forest outperforms the other standard algorithms by showing 91.62% accuracy in crop yield prediction. Further evaluation is done where the attributes that affect the crop yield most are ranked according to their impact based on Mean Absolute Error (MAE). With this, we make a case for the need for a unified approach for crop yield prediction. Key words : Crop Yield, Linear Regression, Random Forest, K Nearest Neighbors (KNN) and Stochastic Gradient Descent (SGD), Machine Learning, Generic Solution. 1. INTRODUCTION Agriculture is one of the important sources of income in India [3]. Farmers grow crops which suits for the environmental condition of their region. Different types of crops are grown in different region because each crop region is defined based on homogeneity in the soil type, climatic conditions, rainfall, farming practices, availability of labor, irrigation facilities and many more [1]. Crop yield prediction is a fundamental agriculture problem to solve. In traditional farming, most of the farmers generally predict the crop yield based on their experience on the specific crop. This ad-hoc approach of crop yield prediction is unreliable and is prone to errors, as the multiple changing environmental factors drastically affect crop yield. In the recent years, various data analytics techniques have been applied in the field of farming to predict the crop yield accurately, and farmers are provided timely advice for planning the future crops boosting up the crop yield as well as quality of crops [2]. Machine Learning algorithms enables handling of vast data, does intelligent analysis by learning patterns in the data and thus guide to improved decision making without human involvement. A significant amount of research work has been carried out in field of Agriculture for accurate prediction of crop yield for different type of crops [3]. Despite the availability of many such models, the farmers in developing nations, such as India, lack the knowledge of these available scientific tools which would guide them for getting best possible farming results. India is a land of diversity in physical features, season, crops, climate and culture, which directly or indirectly influence the annual yield of food crops. For this reason, this paper focuses on crop yield prediction in India. However, the authors believe the same model is applicable globally. This paper provides an analysis of the various solutions available for crop yield prediction technique and attempts to bring out the factors required to develop a generic solution for crop yield prediction. Through a detailed survey of various prediction techniques this work attempts to justify a need for a unified framework for crop yield prediction. Specifically, the major contributions are enlisted below. 1. Study of various solutions proposed/used in crop yield prediction, and the effectiveness of the various parameters that influences their results. 2. Comparative analysis of standard algorithms on crop yield prediction and identifying the most suitable algorithm for a generic set of crops 3. Study of the impact of geographical location on crop yield prediction. 4. Evaluation of various parameters which affects the crop yield and ranking them according to their impact. To pursue this, various standard algorithms and factors/attributes are considered. First, the geography of different states which may have different climatic conditions affects the crop yield. To get more reliable results diverse states of India which are geographically apart, such as Punjab Towards a Unified Approach for Crop Yield Prediction Shilpa Mangesh Pande 1 , Dr. Prem Kumar Ramesh 2 1 Associate Professor, Department of Information Science and Engineering, Research Scholar, Department of Computer Science and Engineering (VTU-RC), CMR Institute of Technology, Bengaluru, India and affiliated to Visvesvaraya Technological University, Belagavi, Karnataka, India, shilpa.p@cmrit.ac.in 1 2 Professor, Department of Computer Science and Engineering, CMR Institute of Technology, Bengaluru, India and affiliated to Visvesvaraya Technological University, Belagavi, Karnataka, India, premkumar.r@cmrit.ac.in 2 ISSN 2347 - 3983 Volume 8. No. 7, July 2020 International Journal of Emerging Trends in Engineering Research Available Online at http://www.warse.org/IJETER/static/pdf/file/ijeter58872020.pdf https://doi.org/10.30534/ijeter/2020/58872020