Satish Kumar Yadav, from the institute of India. D. Pawar, from the institute of India. Latika Yadav , from the institute of India. and Saurabh Tripathi, from the institute of India. wrote a Research Article about, Machine Learning Predicts Thrips Occurrence in Tomatoes Under Weather Variations. Entitled, Predictive Analysis of Occurrence of Thrips in Tomato Subject to Weather Parameters Using Machine Learning Techniques. This research paper published by the International Journal of Biosciences (IJB). an open access scholarly research journal on Biosciences. under the affiliation of the International Network For Natural Sciences | INNSpub. an open access multidisciplinary research journal publisher.
Abstract
Thrips (Thripidae) on tomato (Solanum lycopersicum L.) at Rajendranagar, Andhra Pradesh, India is modelled based on field data sets generated during six kharif seasons [2011-18]. The weather variables considered are maximum & minimum temperature (MaxT & MinT) (0C), morning and evening humidity (RHM & RHE) (%), sunshine hours (SS) (hr/d), wind velocity (Wind) (km/hr), total rainfall (RF) (mm) and rainy days (RD). Thrips incidence was higher during 2012 and lowest in 2014. Correlation analyses significant positive influence of maximum temperature and negative influence of wind of one lags, RHM both current and one lags, rainfall one lag of negative influence on thrips. Machine learning techniques namelyAn empirical comparison of the above models [support vector regression (SVR), random forest (RF) and the other statistical models e.g., multiple linear regression (MLR), ridge regression (RR), least absolute shrinkage and selection operator (LASSO), and elastic net (EN)] is based on root mean square error (RMSE). It is observed that, for thrips, the RMSE values of RF and LASSO are less as compared to other competing models. Diebold-Mariano (D-M) test was applied for comparison of forecasting performance among the applied models. It is observed that, predictive accuracy of RF and LASSO is higher than that of other models.
Introduction
Tomato (Solanum lycopersicum L.) is one of the most popular produced and extensively consumed vegetable crops in the world (Grandillo et al., 1999). It is one of the most important vegetable crops in India that can be eaten raw in salads or as an ingredient in many dishes and in drinks (Alam et al., 2007). Tomatoes and tomato-based foods provide a wide variety of nutrients and many health-related benefits to the body. In regions where it is being cultivated and consumed, it constitutes a very essential part of people’s diet. Tomatoes production accounts for about 4.8 million hectares of harvested land area globally with an estimated production of 165 million tonnes (FAOSTAT, 2017). China leads world tomato production with about 50 million tonnes followed by India with 17.8 million tonnes. Tomato production can serve as a source of income for most rural and peri-urban producers in most developing countries. Despite all the numerous benefits from the crop, many challenges are making its production unprofitable in most developing countries, especially those in Africa. The challenges faced by producers are seen either in production, post-harvest, marketing or a combination of any of them. The purpose of this paper is to look at the postharvest challenges that result in losses and recommend some low-cost intermediate technologies needed to remedy the situation. Accounting for about 8.23% of the total vegetable production in the country. Tremendous progress has been made in tomato production during the past four and half decades. At present, India is the fourth largest producer of tomato, accounting for 6.8% of the world production and the second largest in terms of acreage, accounting for 11.9% of area under tomato in the world. Tomato spotted wilt virus (TSWV) is widely distributed and has caused serious losses in the yield of this and many other crops in Australia, India, Nepal, China, Thailand, and USA. Early infections cause the most severe damage and can lead to total crop loss. Epidemics of insect-transmitted plant viruses in agricultural ecosystems require the interaction of 3 basic components: the host plant of the virus, the insect vector and the plant pathogenic virus. While this triad sounds quite straight forward, the relationships and interactions occurring between and among the basic triad components and the environment are complex and dynamic, frequently defying complete understanding by scientists and agricultural practitioners worldwide. Many plant viruses are transmitted by arthropod vectors (Nault, 1997). TSWV, the type species of the genus Tospovirus, family Bunyaviridae (Murphy et al., 1995), is exclusively transmitted by several thrips species in a propagative manner (German et al., 1992; Ammar, 1994; Goldbach and Peters, 1994). Tomatoes are susceptible to more than 200 diseases. Important achievements in chemical, biological, cultural and genetic control methods have greatly reduced economic losses and sometimes have eliminated them. Viral diseases are a special case since they cannot be controlled by chemical treatments. Crop protection must then rely on genetic resistance or on disease avoidance. TSWV was first reported in India in tomato in 1964 (Todd et al., 1975). The occurrence of TSWV on a legume in India was first recorded in 1968 (Reddy et al., 1968). Thrips (Thysanoptera: Thripidae) cause serious problems in the cultivation of a wide range of greenhouse and field crops. They create major damage on plants by causing reduction in plant growth, deformation of plant organs, and cosmetic damage in the form of silver scars on leaves and flowers. Thrips cause direct damage during feeding, plants should be released by breaking the leaf, fountain and fruits cells, leaving behind silvery patches and fruit sores to reduce plant yields and tomato market shortage (Riley and Pappu, 2004; Staford et al., 2011). And these are dependent on weather conditions (Verhage et al., 2017; Harvey et al., 2018). Thus, there is a need for the development of predictive models for the incidence of pests and diseases that can improve the interpretation of the crop cycle according to the weather, incorporating weather-soil-plant factors (Malau et al., 2018; Badnakhe et al., 2018). Machine learning is a method that works with data analysis and seeks to automate the construction of analytical models (Shekoofa et al., 2014; Li et al., 2016). It is a field of computer science that works with the recognition of patterns using computational learning theory in artificial intelligence (Sahoo et al., 2017). Machine learning algorithms are very promising for faster, more dependent variables and the meteorological elements are the independent variables of the models. Elastic Net (EN), a penalized variable selection approach that combines both ridge penalty and LASSO penalty. Different forecasting techniques e.g., Multiple Linear Regression (MLR); K Neighbors Regressor (KNN); Random Forest Regressor (RFT), and Artificial Neural Networks– Multilayer Perceptron (MLP), EN, LASSO are applied. The ridge method of MLR is utilized. This method avoids poor conditioning of the matrix of the repressor variables, controlling the inflation and the general instability found in least squares estimators. Ridge avoids the multicollinearity problem without having to exclude repressor variables, so it has no information loss.
Reference
Alam T, Tanweer G,
Goyal GK. 2007. Stewart Postharvest Review, Packaging and storage of
tomato puree and paste. Research article. 3(5), 1-8.
Ammar ED. 1994.
Propagative transmission of plant and animal viruses by insects: factors
affecting vector specificity and competence. Advances in Disease
Vector Research 10, 289-331.
Badnakhe MR, Durbha SS,
Jagarlapudi A, Gade RM. 2018. Evaluation of Citrus Gummosis disease
dynamics and predictions with weather and inversion-based leaf optical model.
Computers and Electronics in Agriculture 155, 130-141.
Breiman L. 2001.
Random forests. Machine Learning 45(1), 5–32.
Chatterje S, Hadi
AS. 2012. Regression Analysis by Example, John Wiley & Sons, Inc, New
York.
Chowdappa P. 2010.
Impact of climate change on fungal diseases of Horticultural crops: In:
Challenges of climate change-Indian Horticulture (Eds.: H.P. Singh, J.P. Singh
and S.S. Lal). Westville publishing house, New Delhi. 144-15.
Diebold FX, Mariano RS.
1995. Comparing predictive accuracy. Journal of Business and Economic
Statistics 13, 253-263.
Efron B, Hastie T,
Johnstone I, Tibshirani R. 2004. Least angle regression. The Annals of
Statistics. 32, 407–499.
FAOSTAT 2017.
Global tomato production. Rome, FAO.
German TL, Ullman DE,
Moyer JW. 1992. Tospoviruses: diagnosis, molecular biology, phylogeny, and
vector relationships. Annual Review of Phytopathology 30, 315–348.
German TL, Ullman DE,
Moyerm JW. 1992. Tospoviruses: diagnosis, molecular biology, phylogeny,
and vector relationships. Annual Review of Phytopathology 30, 315–348.
Goldbach R, Peters D. 1994.
Possible cause of the emergence of tospovirus diseases. Seminars in
Virology 5, 113–120.
Grandillo S, Zamir D,
Tanksley SD. 1999. Genetic improvement of
processing tomatoes: A 20 years perspective. Euphytica. 110, 85–97.
Harvey CA,
Saborio-Rodríguez M, Martinez-Rodríguez MR, Viguera B, Chain-Guadarrama A. 2018.
Climate change impacts and adaptation among smallholder farmers in Central
America. Agriculture and Food Security 7(1), 1–20.
Li YH, Xu JY, Tao L, Li
XF, Li S, Zeng X, Prot SVM.-2016. A web-server for machine learning
prediction of protein functional families from sequence irrespective of
similarity. PloS one. 11(8), e0155290.
Liaw A, Wiener M. 2002.
Classification and regression by randomForest. R News. 2, 18-22.
Malau S, Lumbanraja P,
Pandiangan S, Tarigan JR, Tindaon F. 2018. Performance of Coffea arabica L
In Changing Climate of North Sumatra of Indonesia. Scientia Agriculturae
Bohemica 49(4), 340–349. https://doi.org/10.2478/sab-2018-0041.
Murphy FA, Fauquet CM,
Bishop PHL. Ghabrial SA, Jarvis AW, Martelli GP, Mayo MA, Summers MD. 1995.
Virus taxonomy. Sixth report of the international committee on taxonomy of
viruses. Archives of Virology (10), 313–314.
Nault LR. 1997.
Arthropod transmission of plant viruses: a new synthesis. Annals of
Entomological Society of America 90, 521–541.
Paul RK, Vennila S,
Singh N, Chandra P, Yadav SK, Sharma OP, Sharma V, K Nisar S, Bhat MN, Rao MS,
Prabhakar M. 2018. Seasonal Dynamics of Sterility Mosaic of Pigeonpea and
its Prediction using Statistical Models for Banaskantha Region of Gujarat,
India. Journal of the Indian Society of Agricultural Statistics 72, 213-223.
Reddy M, Reddy DVR,
Appa Rao A. 1968. A new record of virus disease on peanut. Plant Disease
Reporter 52, 494-5.
Riley D, Pappu H. 2004.
Tactics for management of thrips (Tysanoptera: Tripidae) and Tomato spotted
wilt virus in tomato. Journal of Economic Entomology 97, 1648–1658.
Sahoo S, Ta R, Elliott
J, Foster I. 2017. Machine learning algorithms for modeling groundwater
level changes in agricultural regions of the US. Water Resources Research 53(5), 3878–3895.
Sakimura K. 1961.
Field observations on the thrips vector species of the tomato spotted wilt
virus in the San Pablo area, California. Plant Disease Reporter 45, 772-776.
Shekoofa A, Emam Y,
Shekoufa N, Ebrahimi M, Ebrahimie E. 2014. Determining the most important
physiological and agronomic traits contributing to maize grain yield through
machine learning algorithms: a new avenue in intelligent agriculture. PloS
one. 9(5), e97288.
Staford CA, Walker GP,
Ullman DE. 2011. Infection with a plant virus modifes vector feeding
behavior. Proceedings of the National Academy of Sciences of the United States
of America 108, 9350–9355, https://doi.org/10.1073/pnas.1100773108.
Tibshirani R. 1996.
Regression Shrinkage and Selection via the Lasso. Journal of the Royal
Statistical Society 58(B), 267–288.
Todd JM, Ponniah S,
Subramanyam CP. 1975. First record of tomato spotted wilt virus from
Nilgiris in India. Madras Agricultural Journal 2, 162-3.
Vapnik VN. 2000.
The Nature of Statistical Learning Theory. Springer- Verlag, New York.
Verhage FYF, Anten NPR,
Sentelhas PC. 2017. Carbon dioxide fertilization off sets negative impacts
of climate change on Arabica coffee yield in Brazil. Clim Chang 144(4), 671–685. https://doi.org/10.Journalpone.0211508.
Zou H, Hastie T. 2005.
Regularization and variable selection via the elastic net. Journal of the Royal
Statistical Society. B67 (2), 301–320.