Journal Volume: 75      No.: 2     Year: 2021
S.No Title Abstract Download
1 Appropriate Statistical Distributions for Yield and Important Auxiliary Characters in Apple
Author: Anju Sharma, P.K. Mahajan, Ashu Chandel, R.K. Gupta and Geeta Verma      Pages: 93-103
The frequency distribution analysis of yield and auxiliary characters utilizing different probability distributions viz., Dagum 3P, Fatigue Life 3P, Gamma 3P, Generalized Gamma 4P, Inverse Gaussian 3P, Log-Logistic 3P, Lognormal 3P and Log-Pearson 3 have been carried out to sample data on different morphological growth characters for selecting the best distribution. The PCA was used to identify the important tree characters which were significantly contributing towards the yield. On analysing the values of different test statistics and based on the scores of goodness of fit tests for different growth variables under study the distribution Dagum 3P was found to be best fitted to spread and yield while Log-Pearson 3 and Inverse Gaussian 3P distributions were most valid fits for number of flowers and number of spurs, respectively. Hence, the results of this study can be used to know the trend and distributional pattern of auxiliary characters of apple crop towards enhancement of yield. Keywords: Principal component analysis, Scree plot, Apple yield, Probability distributions, Goodness of fit tests
2 Information System for Fieldpea Germplasm
Author: Devraj, P.K. Katiya, Rajesh Kumar , Karan Singh, Vikas Deep and Sakshi Mishra      Pages: 105-110
Information System for Fieldpea Germplasm provides a user friendly interface for generating data entry, queries/reports, keep up integrated database for analyzing and interpreting data. The system has been developed on three-tier Client-server architecture using ASP.NET with C# and SQL server 2005. Presently, the system contains information on 480 evaluated for 21 valuable descriptors (12 qualitative traits and 9 quantitative traits) for each attainment for climatic conditions for agriculture in India. Analysis of the data was done on the estimator viz., Mean, Range, Variance, Standard Deviation, Skewness and Kurtosis. System contains two operational sub-systems (viz., Data Management and Report Generation sub-systems). Keywords: Germplasm, Information system, Accessions, Mysql, Descriptors, Client-Server Architecture
3 Selection of Feature Selection Algorithm for Categorization of Research Abstracts in Agricultural Domain
Author: Sreekumar Biswas, Rajni Jain, Sudeep Marwaha, Alka Arora, A.R. Rao and Monendra Grover      Pages: 111-116
Feature selection is one of the most important steps while dealing with classification problems. When it narrows down towards text classification, feature selection becomes indispensable because of the high dimensionality of the feature vector. Feature selection techniques are categorized into a filter, classifier subset evaluation and wrapper. This study presents the empirical results of the comparison of the feature selection for text categorization using research texts in the agricultural domain. The study recommends that the Classifier Subset Evaluator method for feature selection using Na´ve Bays as parameter algorithm and MLP as the classifier is the best framework for categorization of agricultural text documents with 90% accuracy Keywords: Text categorization, Feature selection, Machine learning, Classifier subset evaluation, Wrapper subset evaluation, Artificial intelligence
4 Rescaling Bootstrap Technique for Variance Estimation in Dual Frame Surveys
Author: Rajeev Kumar, Anil R, Tauqueer Ahmad, Ankur Biswas and Pramod Kumar Moury      Pages: 117-125
In a Dual Frame (DF) surveys, set of two frames is used instead of a traditional single frame of sampling units from the target population. Dual frame surveys are applicable in those situations where one frame covers the entire population but very expensive to sample; so an alternate frame may be available that does not cover the entire population but is inexpensive to sample. As Hartley (1962) noted, variance estimation can be more complicated for dual frame surveys than for a single-frame survey. Unbiased variance estimator of parameter of interest is very tedious to obtain for estimator using dual frame surveys. In this article, we propose two rescaling bootstrap variance estimation techniques in dual frame surveys viz. Stratified Rescaling Bootstrap Without Replacement (SRBWO) and Post-stratified Rescaling Bootstrap Without Replacement (PRBWO) methods. Statistical properties of the proposed methods are compared through a simulation study. Simulation results suggest that the proposed SRBWO and PRBWO methods give an unbiased estimate of the variance of the dual frame estimator of population total and the SRBWO method performs better than the PRBWO method. Keywords: Multiple frame surveys, Rescaling bootstrap, Post stratification, Simulation.
5 Need and Impact of Data Analytics in Organic Farming Sector of India
Author: Manavalan      Pages: 127-140
This article mainly focused in bringing out the need and importance of developing and deploying customized Information Technology based data centric Geo-Analytical tools and its related technologies that can monitor and further speed-up the growth of Organic Farming sector of any country. Initially, data pertaining to the Land related Key Indicators of Organic Farming industry of India has been analyzed and its growth status has been brought out since year 2003 to till date. The key indicators such as areal extent of Organic Land, areal extent of Wild collection, number of Organic Producers, Processors, Exporters of India as well as organic market related factors such as Production quantity, Export Volume, Export Value are analyzed in detail. Further to this the significance of deriving as well as relating the outcome of these analysis through a Geographic Information System based Decision Support System (GIS-DSS) is insisted which will further enhance the growth status of Organic Farming sector of India right from taluk, district and at state level. Overall, the need and importance of developing and deploying an advanced ML/DL based integrated GIS based Analytical Decision Support System which can accelerate the growth of Organic Farming sector of India has been brought out. Keywords: Organic farming, Key indicators, Data mining, Geo analytics, IT & ITES
6 Evaluating Machine Learning Approaches for Prediction of Suitable Climatic Conditions for Parthenium hysterophorus (L.) in India
Author: Abhishek Yadav, Yogita Gharde and Sushil Kumar      Pages: 141-145
Parthenium hysterophorus is considered as one of the worst weed of the world as it is difficult to control. Besides extensive documentation on the weed, the details of its current range with intensity level are not fully known in India. Further, Machine Learning algorithms are considered as
7 A Revised Calibration Weight based Ratio Estimator in Two-phase Sampling: A Case when Unit Level Auxiliary Information is Available for the First-phase Sample
Author: Sadikul Islam, Hukum Chandra, U.C. Sud, Pradip Basak, Nirupam Ghosh and P.R. Ojasvi      Pages: 147-156
promising approaches for modelling and predicting the distribution of different species. Hence, the present study aims in evaluating different machine learning approaches (MLA) to find out the best one in predicting the suitable climatic conditions for Parthenium hysterophorus in India. Climatic variables such as mean maximum temperature, mean minimum temperature, relative humidity and rainfall were used as independent variables for prediction. Total of 14 machine learning algorithms were included in the evaluation process. The best algorithm for the prediction was chosen using criteria like percentage of correctly classified instances, Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Root Relative Squared Error (RRSE). Among all machine learning algorithms, J48 was found best for the prediction of suitable climate conditions for establishment of Parthenium in India. This study is helpful in knowing the possible infestation level of Parthenium in a place based on weather conditions which may further be helpful in planning the management strategies timely. Keywords: Machine Learning algorithms, Modelling, Alien invasive weed, Parthenium hysterophorus
8 New D-optimal Covariate Designs in CRD and RCBD set-ups
Author: Hiranmoy Das, Ankita Dutta, Dikeshwar Nishad and Anurup Majumder      Pages: 157-168
This paper proposed a revised calibration weight based finite population ratio estimator under two phase sampling design assuming the condition that two auxiliary variable,correlated to numerator and denominator variable of the ratio were available at unit level for the first-phase sample. But, population level auxiliary variables were not available. In this article, two calibration estimator of finite population ratio were discussed, one consists of ratio of two calibration estimator of total and another, consists of combined common calibration weight based ratio estimator under two-phase sampling design. The results of empirical analysis revealed that combined common calibration weight based ratio estimator was best performing and further both the calibration estimator were outperforming over the existing calibration estimator of Islam et al. (2019). Hence, combined common calibration weight based ratio estimator declared as a proposed revised calibration weight based finite population ratio estimator under two phase sampling. The theoretical expression of variance estimator as well as optimum sample size for minimum variance for fixed cost were also deliberated for the proposed ratio estimator . Keywords: Calibration weight, Cost function, Population ratio, Two-phase sampling.
9 Orthogonally Blocked Second Order Response Surface Designs under Auto-Correlated Errors
Author: C.G. Joshy and N. Balakrishna      Pages: 169-174
One new series of D-optimal covariate designs in Completely Randomized Design (CRD) set-up and two new series of D-optimal covariate designs in Randomized Complete Block Design (RCBD) set-up have been developed in the present study. The methods of construction of D- optimal covariate designs are developed with the help of Special Array (Das et al., 2020). The existence of Hv and Hb-1 can develop c = (v ? 1) number of D-optimal covariates in CRD and RCBD set-ups with the treatment number v for any odd number of replications or blocks, b. Again, if Hv and Hb-r exist, then c = (v ? 1) D-optimal covariates will exist in a RCBD set-up with v number of treatments for b odd number of blocks provided r (> 1) is an odd number and Hr-1 exists. In the Special Array, r is the number of rows (and columns) with all elements zero. The developed optimal covariate designs in this article are not yet available in the existing literature. Keywords: Hadamard matrix, D-optimality, CRD, RCBD
10 Hindi Supplement vol 75 (02)
Author:       Pages: 5