Scribd Logo
Upload

Welcome to Scribd!

  • Prediction of Agricultural Crops Using KNN Algorithm

    Uploaded by

    International Journal of Innovative Science and Research Technology
    21 views3 pages
    Agriculture is full of uncertainty due to climate change, rainfall, soil type and numerous other factors. Crop prediction in agriculture is a very big dilemma and there is huge dataset where farmers find difficult to predict the yield and seed selection

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    Agriculture is full of uncertainty due to climate change, rainfall, soil type and numerous other factors. Crop prediction in agriculture is a very big dilemma and there is huge dataset where farmers find difficult to predict the yield and seed selection

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    21 views3 pages

    Prediction of Agricultural Crops Using KNN Algorithm

    Uploaded by

    International Journal of Innovative Science and Research Technology
    Agriculture is full of uncertainty due to climate change, rainfall, soil type and numerous other factors. Crop prediction in agriculture is a very big dilemma and there is huge dataset where farmers find difficult to predict the yield and seed selection

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    of 3

    Volume 5, Issue 5, May – 2020 International Journal of Innovative Science and Research Technology

    ISSN No:-2456-2165

    Prediction of Agricultural Crops using KNN Algorithm


    H. K. Karthikeya1*, K. Sudarshan2, Disha S. Shetty3
    1,3
    Student, Department of Computer Science & Engineering, Srinivas Institute of Technology, Mangaluru, India
    2
    Associate Professor, Department of Computer Science & Engineering, Srinivas Institute of Technology, Mangaluru, India

    Abstract- Agriculture is full of uncertainty due to climate neighbors to be the prediction query object. As mentioned
    change, rainfall, soil type and numerous other factors. before it can also be used for regression-output, which is the
    Crop prediction in agriculture is a very big dilemma and item's reward .Mostly for distance calculation in K-NN
    there is huge dataset where farmers find difficult to algorithm the metric used is Euclidean distance.
    predict the yield and seed selection. In this current
    situation due to rise in the population the production of crops II. PROBLEM STATEMENT
    and agricultural products needs to be increased
    simultaneously to meet the demands of the people. These In country like India the production of crops are affected by
    problems could be solved using machine learning several factors. Factors like Humidity, temperature, rainfall, soil
    algorithms and this paper focuses on these solutions. The type play a vital role in crop prediction, and factors like these differ
    real time environmental parameters like soil type, rainfall, by large with respect to region. In India farmers majorly still rely on
    humidity etc of Mangalore, Kodagu, Kasaragod and some traditional techniques inherited from their forefathers. These
    other districts of Karnataka state are collected and crop techniques would work earlier when the climate was much
    prediction is done along with the accuracy for the crops is healthier and predictable. Now with factors like global warming
    done with the help of K-NN algorithm. and pollution affecting the environment people have to be smart
    and start utilizing modern techniques. It is time to analyze large set
    Keywords:- Agriculture, Crop-Prediction, K-Nearest of data and come up with a system that can provide sufficient
    Neighbor . information regarding crop yield. The new age methodology
    requires large structured data sets and an algorithm capable of
    I. INTRODUCTION providing solution using the provided datasets.

    India is an agricultural country. India’s economy is III. METHODOLOGY


    determined by agricultural products export and import.
    Agriculture is one of the important aspects of Indian A. Dataset Collection
    economy. Due to uncertainty in the crop yield there is a great When implementing an accurate prediction model it
    fall in the economic status. The major crops of India are Rice, might not be sufficient to just consider one or two parameters.
    Wheat, Pulses and Grains. Day by day the population of India Data about Rainfall, temperature, humidity and various other
    is growing and the crops productivity need to be increased to factors are collected and analyzed. This analysis will be fed to
    feed the population. One of the best ways of predicting the prediction model.
    unknown values is by use of machine learning algorithms.
    This work intends to develop crop prediction model using
    machine learning. The application intends to predict crop yield
    so it could help farmer to choose best seeds for plantation.
    There are plenty of ML algorithms which could be used,
    algorithms like Regression analysis, Support Vector Machine,
    Neural Networks, K-Nearest Neighbor (K-NN) can be
    utilized. In this work we discuss about K-NN. The k-nearest
    neighbors (KNN) algorithm is a simple, supervised machine learning
    algorithm that can be used to solve both classification and regression Fig 1:- Flow graph of the methodology Data Collection
    problems. It’s easy to implement and understand, but has a major
    drawback of becoming significantly slows as the size of that data in Here we gather information from several sources and
    use grows. Here objective is to use a model where information construct datasets. Plenty of online portals like Raitha-mithra,
    focuses are clustered in a few groups in order to predict the karnataka.gov.in and Data.gov.in [1] are available for information
    classification of another instance. K-NN works based on collection. Annual crop report of each crop is collected
    minimum distance from query instance to the training samples Collecting previous crop history data from places like
    to determine the k-nearest neighbors. Then we collect k- Mangalore, Kodagu, Kasaragod, Mysore, Davangere, Hassan,
    nearest neighbors, we take simple majority of these k-nearest Shivamogga, Chikkamagalur which belongs to Karnataka State.

    IJISRT20MAY722 www.ijisrt.com 1422


    Volume 5, Issue 5, May – 2020 International Journal of Innovative Science and Research Technology
    ISSN No:-2456-2165
    Collecting data related to crops like Coconut, Cardamom, IV. TESTING AND ANALYSIS
    Coffee, Areca nut, Ginger, Tea, Paddy, Ground nut, Black
    gram, Cashew, Pepper are the crops which are commonly The purpose of the test was to find the workings of K-
    grown in these regions. We also collect data related to NN algorithm and how will it predict the yield when three
    Rainfall. Humidity, Soil type, Irrigation type, Previous Yields, parameters were given as input. Input data is given as follows
    Location, Price, Year, type of crop, Crop diseases and its Location: Mangalore, Soil-Type: Coastal alluvial and Area:
    symptoms 1395 cents. The system predicted Coconut and Cocoa as two
    potential crops with the accuracy of 63.63%. While testing for
    B. K-NN Algorithm Kodagu district the where soil type is Laterite soil and the area
    The k-nearest neighbor (k-NN) method is a data mining given was 1395 cents the system predicted Cardamom and
    technique considered to be among the top five techniques for Pepper as two potential crops and noted the accuracy of
    data mining. In this, we consider each of the characteristics in 56.66%
    our training set as a different dimension in some space, and
    take the value an observation has for this characteristic to be
    its coordinate in that dimension, so getting a set of points in
    space. We can then consider the similarity of two points to be
    the distance between them in this space under some
    appropriate metric. The way in which the algorithm decides
    which of the points from the training set are similar enough to
    be considered when choosing the class to predict for a new
    observation is to pick the k closest data points to the new
    observation, and to take the most common class among these.
    This is why it is called the k Nearest Neighbors algorithm. The
    implementation of algorithm can be noted as below :

    1. Load the data


    2. Initialize K to your chosen number of neighbors
    3. For each example in the data
     Calculate the distance between the query example and the Fig 2:- Accuracy of KNN Algorithm for Mangalore Region
    current example from the data.
     Add the distance and the index of to an ordered collection.
    4. Sort the ordered collection of distances and indices from
    smallest to largest (in ascending order) by the distances
    5. Pick the first K entries from the sorted collection
    6. Get the labels of the selected K entries
    7. If regression, return the mean of the K labels
    8. If classification, return the mode of the K labels

    C. Prediction of Crop Yield through KNN


    Here we consider parameters like humidity, rainfall, soil
    type, area etc. We have assigned location, area, soil type as
    input parameters although other parameters may also be
    considered. The crop yield which is an unknown value can be
    predicted using the values of the nearest known neighbors.
    This is possible by calculation Euclidian distance between
    those points. Thus we will be able to predict crop yield for the
    given input parameters. The calculation of distance between
    points in a feature space, different distance functions could be Fig 3:- Accuracy of KNN Algorithm for Kodagu Region
    used, in which the Euclidean distance function is the most
    commonly used one. Say p and q are represented as feature The system helps in avoiding the use of sensors and
    vectors. To measure the distance between p and q, the reduces unnecessary cost. This system results in efficient
    Euclidean metric is generally used by if a = (a1, a2) and b = usage of time and cost. A key aspect of Crop Prediction is to
    (b1,b2) then the distance is given by: identify a suitable crop quickly and suggest the farmer as to
    which crop to grow. Our system helps in gathering all
    necessary information and giving a model of output which not
    d(a, b) = √(𝑏1 − 𝑎1)2 + (𝑏2 − 𝑎2)2 only increases current economical gain but also safeguards
    future profitability. The accuracy part of the system is noted as

    IJISRT20MAY722 www.ijisrt.com 1423


    Volume 5, Issue 5, May – 2020 International Journal of Innovative Science and Research Technology
    ISSN No:-2456-2165
    decent but can be made more accurate with increase in the
    efficiency.

    V. CONCLUSION

    The implementation of the system was to learn about


    crops and agriculture and find an efficient way of harvesting.
    The study focuses on the agricultural datasets obtained from
    various portals belonging to some districts of Karnataka State.
    Datasets ordered in well structured manner. K-NN algorithm
    is used for the prediction model and crop yield prediction and
    its accuracy is obtained. The future is bright for the
    implementation of machine learning algorithms in the field of
    crop production and we hope to implement more advanced
    algorithms so that the system becomes more efficient, we hope
    to make system prediction more stable and obtain high
    accuracy with the help of more datasets and advanced
    algorithms.

    REFRENCES

    [1]. P. Vinciya “Agriculture Analysis for Next Generation


    High Tech Farming in Data Mining”, Anna University,
    Trichy, Tamilnadu, India, 5 May,2016.
    [2]. M. A. Jayaram and Netra Marad,” Fuzzy Inference
    Systems for Crop Yield Prediction”, Journal of
    Intelligent Systems, 2012,21(4),pp.363-372.
    [3]. Sk Al Zaminur Rahman, Kaushik Chandra Mitra, Soil
    Classification using Machine Learning Methods and
    Crop Suggestion Based on Soil Series,2018 21st
    International Conference of Computer and Information
    Technology (ICCIT), 21-23 December 2018.
    [4]. Hart, Peter E. (1968). "The Condensed Nearest
    Neighbor Rule". IEEE Transactions on Information
    Theory. 18: 515–516. doi:10.1109/TIT.1968.1054155
    [5]. http://agricoop.nic.in/sites/default/files/Annual_rpt_2016
    17_E.pdf.

    IJISRT20MAY722 www.ijisrt.com 1424

    You might also like