111
CROP YIELD PREDICTION
Mrs.K.R.Sri Preethaa M.E.,
Assistant Professor (Sr.G), Department of Computer Science and Engineering
S.Nishanthini, D.Santhiya, K.Vani Shree
UG Students, Department of Computer Science and Engineering
KPR Institute of Engineering and Technology,
Arasur, Coimbatore.
ABSTRACT:
Agriculture is the backbone of our country and the economic growth of our nation. This paper discusses the aim to increase the net yield rate of the crop, based on the parameters related to soil and atmosphere. This paper helps us to predict the crop yield and suggest the best crop thereby improving the quality and profitability of the agricultural sector by processing the datasets. The parameters included in the datasets are soil type, temperature, humidity, water level, spacing, depth, soil ph, season, fertilizer and months. This prediction will help the farmers to choose whether the particular crop is suitable for that specific soil. This prediction can be carried out by using the Bayesian algorithm where high accuracy and speed can be achieved.
Keywords: Crop yield, Data sets, Analysis and Prediction, Bayesian algorithm.
INTRODUCTION:
From ancient period, agriculture is considered as the main and the foremost culture practiced in India. Ancient people cultivate the crops in their own land and so they have been accommodated to their needs. Therefore, the natural crops are cultivated and is been used by many creatures such as human beings, animals and birds. The greenish goods produced in the land which have been taken by the creatures leads to a healthy and welfare life. Since the invention of new innovative technologies and techniques the agriculture field is slowly degrading. Due to these abundant invention people are been concentrated on cultivating artificial products that is hybrid products where there leads to an unhealthy life. Nowadays, modern people don‟t have an awareness about the cultivation of the crops in a right time and at
a right place. Because of these cultivating techniques the seasonal climatic conditions are also being changed against the fundamental assets like soil, water and air which leads to insecurity of food.
112
analysing there are various algorithms and here we use “Bayesian Algorithm” where accuracy can also be measured. For analysing and predicting the suitable crop for a particular soil the data sets are being collected based on weather, season, soil PH and water level. By Bayesian Algorithm, the probabilistic conditions are calculated and the prediction is done.
RELATED WORK:
AGRICULTURAL
RECOMMENDER
USING
DATA
MINING TECHNIQUES
Using data mining or KDD (Knowledge Discovery in Database), the data sets are extracted from large amount of data. For data to be optimized genetic algorithm is used where chromosomes, mutation and recombination are considered. The frequent item sets are collected using ARM (Association Rule Mining) where two things focussed such as support and confidence.
Support- frequent item are generated by a certain rule.
Confidence- mainly focuses on rule generation, ie. how frequently certain items appear for particular transaction.
Thus, an optimized result can be gained by using these techniques(genetic and ARM). Till now, the crop yield can be predicted using soil type and climatic condition. Besides this fertilizers and pesticide can also be used to predict the production of the crop yield exactly. Thus by these economical status in agricultural sector can be increased.
SYSTEM
FOR
AGRICULTURE
RECOMMENDATION
USING
DATA MINING
In addition to ARM and Genetic
algorithm (AGRICULTURAL
RECOMMENDER USING DATA MINING TECHNIQUES), the web based system can be used to extract the required result for predicting the crop. In this paper, the prediction is mainly done for the purpose of sowing.
Steps in web based system,
1. Registration
2. Extracting the suitable crop
3. Messaging system to customer
Mathematical model can also be used to calculate the yield.
STUDY
OF
CLUSTERING
TECHNIQUES
FOR
CROP
PREDICTION
The prediction of crop yield is important in agriculture. The quality of crop will be affected due to the unavailability of proper knowledge about the crops. Several clustering methods are used such as model based, hierarchical, grid based, constrained based and partitioning to gain useful information in mining. The comparative study about Bee-Hive and improved K-means algorithm is done which leads to performance evaluation.
In this paper, two methods have been categorized such as predictive and descriptive methods where clustering plays a vital role for finding the data information and pattern recogonition. Some clustering algorithms are,
113
A large number of data sets cannot be handled using SVM, therefore CB-SVM have been proposed where large data sets can be handled.
2. Constrained K-means Algorithm
The constrained based and partitioning methods are merged in this algorithm.
1. SWK K-means Algorithm
a. In this algorithm information such as kernel matrix, number of clusters, weight for each point, stopping criteria, penalty term parameters are added.
b. For better performance good quality clusters are used for huge data sets.
2. Expectation Maximization (EM) Algorithm
The parameters are refined based on expectation and maximization steps.
APPLYING
DATA
MINING
TECHNIQUES IN THE FIELD OF
AGRICULTURE
AND
ALLIED
SCIENCES
Data mining is the approach for predicting the crop in the field of agriculture. Some of the techniques used for prediction such as ID3 algorithms, the k-means, the k-nearest neighbour, artificial neural networks and support vector machines. It explores the applications of data mining techniques in the field of agriculture and allied sciences. In this paper, converting the data into useful information and knowledge. This information is useful for business management, production control, market analysis, to engineering design and science. Many types of functionalities are used such as data collection and database
creation , data management(including data storage and retrieval and data transaction processing) and data analysis and understanding(involving data warehousing and data mining).
For instance, the data collection and database creation mechanisms served as a 3effective mechanism for data storage and retrieval and transaction processing. Data mining techniques are used to study of soil characteristics. For example, the k-means approach is used for classifying soils in the combination with GPS-based technologies and plants same as support vector machines (SVM) to classify crops.ID3 algorithm is used to generate the decision tree from a datasets. The k-nearest neighbour (KNN) stores all available cases and classifies new cases based on a similarity measure and also used in statistical estimation and pattern recognition. The multidisciplinary approach of integrating computer science with agriculture is helpful in forecasting agriculture crops effectively.
PROPOSED ARCHITECTURE:
Collection of
Agricultural Datasets
Selection of parameters
Prediction based on parameters
114
In this proposed system, the datasets are collected and refined based on the commonality. The input parameters are given. By analysing and predicting using Bayesian algorithm, the result are produced and some suggestions are given.
PROPOSED WORK:
Step 1:
The datasets have been collected and refined based on commonality uses such as soil type, temperature, humidity, water level, spacing ,depth, soil ph, season, fertilizer and months. These data sets are entered into the database using mysql queries. From these parameters name of the crop and net yield rate of the crop can be predicted.
Step 2:
Based on various analyses the parameters soil type and temperature are taken as input and prediction have been undertaken. The attribute soil type specifies the type of soil in a particular region such as Alluvial, Loamy, Black soil, Clay and Red and the attribute temperature specifies the water content available in the soil.
Step 3:
By using Bayesian algorithm, the particular crop has been analysed and predicted by taking various parameters into an account such as soil type and temperature.
How to predict and analyse?
1. I
dentify the tuples with attributes in the datasets.
2. S
ummarize the properties in the dataset and calculate the probabilities and make predictions. After that, use the summarize dataset to generate a single prediction.
3. T
he classifier will predict the particular attributes needed with the conditional probability theorem.
4.
If the needed attributes are not available in the data sets then the conditional probability theorem predicts the suitable output and will be analyzed.
Step 4:
By analysing and predicting the crop name and net yield rate of particular crop can be found out and also by giving the various suggestions such as spacing, depth and fertilizer for particular crop. This helps the farmers to take the correct decision to sow the crops such that yield rate can be increased.
PROCEDURE:
According to this algorithm, convert the collected data sets into csv file format and load those data sets. Split the loaded data sets into two sets such as „training data‟ and „test data‟ in the split ratio of either 67 percentages or 33 percentages that is 0.67 or 0.33. Separate the training data by class values so that the attribute map to a suitable values and stored in appropriate list. Then calculate „Mean‟ and „Standard Deviation‟ for needed tuple and then summarize the data sets.
115
can be predicted by comparing the resultant class value with the test data set. The accuracy can range from 0% to 100%.
CONCLUSION:
This paper may improve the net yield rate of the crop by comparing and analysing the parameters which is in the datasets. Analysis and Prediction can be done by using the Bayesian algorithm where accuracy can also be predicted. This makes the farmers to take the right decision for right crop such that the agricultural sector will be developed by innovative ideas.
FUTURE WORK:
Now-a-days diseases are affecting drastically to everyone in this world. We cannot predict which type of crops are affecting by what type of diseases.The future study is to enhance the type of diseases affected by suitable crop and also suggesting to use a type of pesticide in order to overcome from those diseases.
REFERENCES:
[1] Rakesh Kumar, M.P. Singh, Prabhat Kumar and J.P. Singh,” Crop Selection Method to Maximize Crop Yield Rate using Machine Learning Technique”,Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, T.N., India. 6 - 8 May 2015. pp.138-145.
[2] Ramya M, Chetan Balaji, Girish L ,” Environment Change Prediction to Adapt Climate-Smart Agriculture Using Big Data Analytics”, International Journal of Advanced Research in Computer Engineering & Technology (IJARCET) Volume 4 Issue 5, May 2015 .
[3] Diego Fabian Pajarito Grajales, Geidy Jhoana Asprilla Mosquera, Fabian Mejia, Leonardo Cardona Piedrahita, Cesar Basurto, “Crop-Planning, Making Smarter Agriculture With Climate Data”.
[4] Hongyu Guo · Herna L. Viktor, ” Multirelational classification: a multiple view approach”, 26 February 2008.
[5] Yethiraj N G , ” Applying data mining techniques in the field of Agriculture and allied sciences”, Vol 01, Issue 02, December 2012.
[6] Rumelhart DE, Hinton GE, Williams RJ, ”Learning internal repre- sentations by error propagation”. vol. 1, chapter 8. The MIT Press, Cambridge, MA (USA), pp:418-362, 1986.
[7] Liu J, Goering CE, Tian L, ”Neural network for setting target corn yields”. T ASAE 44(3): 705-713, 2001.
[8] Drummond ST, Sudduth KA, Joshi A, Birrel SJ, Kitchen NR, ”Statistical and neural methods for sitespecific yield prediction”. T ASABE 46 (1): 5-14, 2003.
[9] Safa B, Khalili A, Teshnehlab M, Liaghat A, ”Artificial neural networks application to predict wheat yield using climatic data. Proc”. 20th Int. Conf. on IIPS, Jan. 10-15, Iranian Meteorological Organization, pp: 1-39, 2004.