Академический Документы
Профессиональный Документы
Культура Документы
KeyWords: Map-Reduce,HDFS,K-Nearest
Neighbour,Predictive Analysis.
1.INTRODUCTION
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 03 | Mar 2019 www.irjet.net p-ISSN: 2395-0072
Present sites and application give ,so that it can be easy to understand.We select the month
suggestions only for identifying, in which we plan to sow the crop and then we provide the
Crop yield production rate previous year information analyse the data sets of that
Soil Conditions particular month and then classify the data based
Diseases of Crops ondisease and data extracted from the classifier and finally
Weather Conditions predict the soil and crop. The prediction of the soil is
Online Fertilizers shopping etc., (Study) represented with a pie-chart with the respective
Some of issues arising in a existing system are, percentages of the prediction of the crop. Based on the
predictions, the crop and soil are divided in to five
High use of fertilisers categories such as “very good”, “good”, “average”, “bad” and
Susceptibility to pests “very bad”. [5]
Environmental pollution
Loss of biodiversity IV.IMPLEMENTATION
Monoculture policy will be increased if the proper 1.Dataset collection: Here in this process, we collect all
suggestive measures not discussed between the harvesters, the required dataset. Regarding the datasets the initial
hence generating the waste of yielded crops. Only
description and the point to be remembered is about the
Experienced farmers see benefit through planting different
crop variety, that is they sow a high yield variety of crop attributes. The dataset regarding the attributes which
and see income but others efforts gone in vain this is the suits the project must be analysed such that the entire
main problem using the existing Data mining system. results depend on the dataset collected and attributes
containing in it.[2] The data set collected are
There exist numerous methods and proposed
mechanism/models for the prediction of crop yield with crops vs. seasons
innovative ways of analysing and classifying datasets. But crops vs. price(various years)
they hardly discuss the issues and methods of handling
large and complex datasets are, Hence classifying large 2.Map reduction: Map reduction technique in the
datasets remains a very difficult and complicated task with Hadoop tool is used to get the data that’s only required.
an additional expectation of enhanced performance makes Map reduction technique reduces the amount of data
it more challenging. Soft computing and advanced technical
needed to be processed by the classification algorithm is
methods have been applied in the field of farming such as
artificial neural networks, the k nearest neighbour, the k- reduced.
means, support vector machines and ID3 algorithms.The
application of Data mining techniques in is a moderately
new approach and provides the prediction of agricultural
crops. India has the ability to achieve extraordinary
increase in the crop yield production with the help of
expansion of irrigation and technological innovation in
agriculture. There are multiple strategies which can be
adopted to improve the number and quality of crops. [5]
III.PROPOSED SYSTEM
Instead of using the data mining techniques we will use
the Map reduction Techniques of Big Data Analysis. The Big
data Analytics provides following advantages compared to
Fig 2: Map-Reduce operation on Crop Demand
data mining techniques.[2]
Faster and Better decision making by It is a method of calculating demand of common crops
using distribute solutions. from input files on HDFS. The information saved in HDFS
Extensibility and More Reliable way of should split into varied partitions and mapped to workers.
analyzing data. The process of mapper phase is to take the input files and
map it into intermediate < key, value > pairs of shuffled
We create an analysis site which perform Hadoop
and deposited into local disks, in which each file holds
based functions and predict the demanded crop which is
records with one particular key. When mapping process is
not grown in order to maintain Sustainable Growth to
over, reducer phase recover the mapping files from
Farmers Strain. We collect the Data from Agri based
workers of remote device and begins the reducing process
government sites of recently grown and predict the crops
and eventually stores the top outcome to output files on
which is required in nearest future, so that farmers can be
HDFS. [2]
saved from getting loss. And convert it to statistical format
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 2
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 03 | Mar 2019 www.irjet.net p-ISSN: 2395-0072
OS Ubuntu 18.04
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 03 | Mar 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 4