Forecasting with Artificial Neural Networks:
Literature Review (Survey)
Vikas Lamba
Computer Science and Engineering Department
Jaipur National University, Jaipur, Rajasthan, India
ABSTRACT:-This paper shows the literature survey of forecasting of wheat yield in the field of
Agricultural Science. Paper shows all the past development research in this area like all
forecasting models those used in Crop forecasting. The major model areas are like Statistical,
Metrological, Simulation, Agronomic, Remote Satellite Sensed, Synthetic and Mathematical all
these models are in the field of Agricultural Prediction. This paper is a combination of all these
models researches.
INTRODUCTION:
Recent research activities in artificial neural net-works (ANNs) have shown that ANNs have
powerful pattern classification and pattern recognition capabilities. Inspired by biological
systems, particularly by research into the human brain, ANNs are able to learn from and
generalize from experience. Currently, ANNs are being used for a wide variety of tasks in many
different fields of business, industry and science (Widrow et al., 1994). One major application
area of ANNs is forecasting (Sharda, 1994). ANNs provide an attractive alternative tool for both
forecasting researchers and practitioners. Several distinguishing features of ANNs make them
valuable and attractive for a forecasting task.[1]
First, as opposed to the traditional model-based methods, ANNs are data-driven self- adaptive
methods in that there are few a priori assumptions about the models for problems under study.
They learn from examples and capture subtle functional relationships among the data even if the
underlying relationships are unknown or hard to describe. Thus ANNs are well suited for
problems whose solutions require knowledge that is difficult to specify but for which there are
enough data or observations. [2] In this sense they can be treated as one of the multivariate
nonlinear nonparametric statistical methods (White, 1989; Ripley, 1993; Cheng and
Titterington, 1994). This modeling approach with the ability to learn from experience is very
useful for many practical problems since it is often easier to have data than to have good
theoretical guesses about the underlying laws governing the systems from which data are
generated. [3][4] The problem with the data-driven modeling approach is that the underlying
rules are not always evident and observations are often masked by noise. It nevertheless
provides a practical and, in some situations, the only feasible way to solve real-world problems.
Second, ANNs can generalize. After learning the data presented to them (a sample),
ANNs can often correctly infer the unseen part of a population even if the sample data contain
noisy information. As forecasting is performed via prediction of future behavior (the unseen part)
from examples of past behavior, it is an ideal application area for neural networks, at least in
principle.[5]
Third, ANNs are universal functional approximations. It has been shown that a network can
approximate any continuous function to any desired accuracy (Irie and Miyake, 1988; Hornik et
al., 1989; Cybenko, 1989; Funahashi, 1989; Hornik, 1991, 1993). [6] ANNs have more general
and flexible functional forms than the traditional statistical methods can effectively deal with.
Any forecasting model assumes that there exists an underlying (known or unknown) relationship
between the inputs (the past values of the time series and/or other relevant variables) and the
outputs (the future values). Frequently, traditional statistical forecasting models have limitations
in estimating this underlying function due to the complexity of the real system. ANNs can be a
good alternative method to identify this function.[7]
Finally, ANNs are nonlinear. Forecasting has long been the domain of linear statistics. The
traditional approaches to time series prediction, such as the Box-Jenkins or ARIMA method
(Box and Jenkins, 1976; Pankratz, 1983), assume that the time series under study are generated
from linear processes. Linear models have advantages in that they can be understood and
analyzed in great detail, and they are easy to explain and implement. However, they may be
totally inappropriate if the underlying mechanism is nonlinear. It is unreasonable to assume a
priori that a particular realization of a given time series is generated by a linear process. In fact,
real world systems are often nonlinear (Granger and Terasvirta, 1993). During the last decade,
several nonlinear time series models such as the bilinear model (Granger and Anderson, 1978),
the threshold autoregressive (TAR) model (Tong and Lim, 1980), and the auto- regressive
conditional hetero sciatic (ARCH) model (Engle, 1982) have been developed. However, these
nonlinear models are still limited in that an explicit relationship for the data series at hand has to
be hypothesized with little knowledge of the underlying law. In fact, the formulation of a
nonlinear model to an articular data set is a very difficult task since there are too many possible
nonlinear patterns and a pre specified nonlinear model may not be general enough to capture all
the important features. [3][8][9]
Artificial neural networks, which are nonlinear data-driven approaches as opposed to the
above model-based nonlinear methods, are capable of performing nonlinear modeling without a
priori knowledge about the relationships between input and output variables. Thus they are a
more general and flexible modeling tool for forecasting. The idea of using ANNs for forecasting
is not new. The first application dates back to 1964. Hu (1964), in his thesis, uses the Widrow’s
adaptive linear network to weather forecasting. Due to the lack of a training algorithm for
general multi-layer networks at the time, the research was quite limited. It is not until 1986 when
the bckpropagation algorithm was introduced (Rumelhart et al., 1986b) that there had been much
development in the use of ANNs for forecasting. Werbos (1974), (1988) first formulates the back
propagation and finds that ANNs trained with back propagation outperform the tradi- tional
statistical methods such as regression and Box-Jenkins approaches. [1][11]
Lapedes and Farber (1987) conduct a simulated study and conclude that ANNs can be
used for modeling and forecasting nonlinear time series.Weigend et al. (1990), (1992); Cottrell et
al. (1995) address the issue of network structure for forecasting real-world time series. Tang et
al. (1991), Sharda and Patil (1992), and Tang and Fishwick(1993), among others, report results
of several forecasting comparisons between Box-Jenkins and ANN models. [12]
Heping Pan, Chandima Tilakaratne, John Yearwood, 2005, presents a computational approach
for predicting the Australian stock market index –AORD using multi-layer feed-forward neural
networks from the time series data of AORD and various interrelated markets. This effort aims to
discover an effective neural network or a set of adaptive neural networks for this prediction
purpose. [13]Christopher Gan,Visit Limsombunchai, 2005, interest in applying artificial neural
networks (ANN) to analyze consumer behavior and to model the consumer decision-making
process. KOŠCAK et al., 2009, have compared common meteorological forecasting method with
ANN and he found the performance of ANN with high accuracy Mahdi Pakdaman Naeini,
Hamidreza Taremian, Homa Baradaran Hashemi, 2010, two kinds of neural networks, a feed
forward multilayer Perceptron (MLP) and an Elman recurrent network, are used to predict a
company’s stock value based on its stock share value history. [15] The experimental results show
that the application of MLP neural network is more promising in predicting stock value changes
rather than Elman recurrent network and linear regression method.
Nekoukar et al., 2010, have used radial basis function neural network for financial timeseries forecasting, and the result of their experiment shows the feasibility and effectiveness.
Geetha and Selvaraj, 2011, have predicted Rainfall in Chennai using back propagation neural
network model, by their research the mean monthly rainfall is predicted using ANN model.
Jyothi Patil, V.D.Mytri, 2012, Researchers have attempted to comprehend the pest population
dynamics by applying analytical and other techniques on pest surveillance data sets. In this
paper, an intelligent system for effectual prediction of pest population dynamics of Thrips Tabaci
Linde (Thrips) on cotton (Gossypium Arboreum) crop is presented. [17]
The feed forward Multi-Layer Perception (MLP) Neural Network with back propagation
training algorithm is employed in the design of the intelligent system. The neural network is
trained and tested with the data prepared. [18] The experimental results portray the effectiveness
of the proposed system in predicting pest population dynamics of Trips’ on cotton crop.
Moreover, a comparative analysis is performed between the proposed system and two of the
existing works.
Jyothi Patil, V.D.Mytri, 2012 Researchers have attempted to comprehend
the pest population dynamics by applying analytical and other techniques on pest surveillance
data sets. In this paper, An intelligent system for effectual prediction of pest population dynamics
of Thrips Tabaci Linde (Thrips) on cotton (Gossypium Arboreum) crop is presented. The feed
forward Multi-Layer Perceptron (MLP) Neural Network with backpropagation training
algorithm is employed in the design of the intelligent system. [19]
The neural network is trained and tested with the data prepared. The experimental results
portray the effectiveness of the proposed system in predicting pest population dynamics of
Thrips on cotton crop. Moreover, a comparative analysis is performed between the proposed
system and two of the existing works. Prakash Ramani, Dr. P.D. Murarka, 2013 proposed a stock
price prediction model using multi-layer feed forward Artificial Neural Network (ANN). In this
model we have used backpropagtion algorithm. As the closing price of any stock already covers
other attributes of the company, we have used historical stock prices (closing) for training the
network. [20]
Dase R.K. and Pawar D.D. in predicated stock rate because it is a challenging and daunting task
to find out which is more effective and accurate method so that a buy or sell signal can be
enervated for given stocks. Predicting stock index with traditional time series analysis has proven
to be difficult an artificial neural network may be suitable for the task. [21] Lubomir Macku and
David Samek, 2013, contribution studies prediction of the given semibatch reactor using
multilayer feed-forward neural networks. The two prediction approaches are tested – signal
prediction approach and system prediction methodology. The first approach is commonly applied
in time series prediction, while the input-output models in the second methodology are used for
example in the control tasks. Furthermore, the resulting predictor is used for the model predictive
control of the reactor in order to test performance of the developed method.(Artificial neural
networks are commonly used in various fields, for example weather forecasting time series
prediction of financial data, biology and medicine, power engineering and process control. There
is lot of types of artificial neural networks, but not all of them are usable for prediction.
The most common are multilayer feed-forward neural networks. Fairly wide group of
artificial neural networks belongs to recurrent neural networks. Very popular due to their fast
training are radial basis function neural networks. [1][22]Alshayea and Elrefae use generalized
regression neural network for prediction of Spanish banks data) develop Neural Network Model
for energy consumption and analyze the performance model. Energy consumption prediction
which focuses on structures and the parameters used in developing Neural Network models
proposed neural network energy prediction model is able to demonstrate an adequate
performance with least Root Mean Square Error.
REFERENCES:
[1] Jyothi Patil, V.D.Mytri “A Prediction Model for Population Dynamics of Cotton Pest (Trips’
tobacco Lined) using Multilayer-Perception Neural Network” International Journal of Computer
Applications (0975 – 8887) Volume 67– No.4, April 2013.
[2] Guoqiang Zhang, B. Eddy Patuwo, Michael Y. Hu “Forecasting with artificial neural
networks: The state of the art” International Journal of Forecasting 14 (1998) 35–62.
[3] Reza Ghodsi, Ruzbeh Mirabdollah Yani, Rana Jalali, Mahsa Ruzbahman “ Predicting Wheat
Production in Iran Using an Artificial Neural Networks Approach” International Journal of
Academic Research in Business and Social Sciences February 2012, Vol. 2, No. 2 ISSN: 22226990.
[4] Vida Varahrami “Good Prediction of Gas Price between MLFF and GMDH Neural network”
International Journal of Finance and Accounting 2012, 1(3): 23-27
[5] Adesh Kumar Pandey, A.K Sinha, and V.K Srivastava”A Comparative Study of NeuralNetwork & Fuzzy Time Series Forecasting Techniques – Case Study: Wheat Production
Forecasting” IJCSNS International Journal of Computer Science and Network Security, VOL.8
No.9, September 2008.
[6] Yue Ma, Yu Chang, Chunyu Xia “Applied Research on Stock Forecasting Model Based on
BP neural network” International Conference on Electronic & Mechanical Engineering and
Information Technology 2011.
[7] Arti R. Naik, Prof. S.K.Pathan ”Wether classification and Forecasting using Back
Propagation Feed-Forward Neural Network” International Journal of Scientific and Research
Publications, Volume 2, Issue 12, December 2012 1 ISSN 2250-3153.
[8] Christopher Gan, Visit Limsombunchai, Mike Clemes, Amy Weng” Consumer Choice
Prediction: Artificial Neural Networks versus Logistic Models” Commerce Division Discussion
Paper No. 104, July 2005.
[9] L. Chenl, M. R. Smith, Y. Hui “Comparing the Performance of Two Feedforward Neural
Network Training Algorithms in MRI Reconstruction” IEEE 1996.
[10] Sean Stijven, Wouter Minnebo, Katya Vladislavleva “Separating the Wheat from the Chaff:
On Feature Selection and Feature Importance in Regression Random Forests and Symbolic
Regression” July 12–16, 2011, Dublin, Ireland ACM 978-1-4503-0690.
[11] Muhammad Awais, Shafay Shamail. Nisar Ahmed. and Saman Shahid “A Comparative
Study of Feed Forward Neural Networks and Radial Basis Neural Networks for Modeling
Tokamak Fusion Process” Proceedings IEEE INMlC 2003.
[12] Mahdi Pakdaman Naeini, Hamidreza Taremian, Homa Baradaran Hashemi “Stock Market
Value Prediction Using Neural Networks” IEEE 2010.
[13] Kanna Bhaskar, S. N. Singh “AWNN-Assisted Wind Power Forecasting Using FeedForward Neural Network” IEEE Transactions on Sustainable Energy, vol. 3, no. 2, April 2012.
[14] Bian Feng-lan, Cai Hai-quan “The Prediction Model of Highway Network Scale Based on
Traffic Demand” 2013 Fifth Conference on Measuring Technology and Mechatronics
Automation.
[15] Seyed Hadi Hosseini, Behzad Moshiri, Ashkan Rahimi-Kian, Babak N. Araabi, “ShortTerm Traffic Flow Forecasting by Mutual Information and Artificial Neural Networks” 2012
IEEE.
[16] Zhongbo Liu, Zhaosheng Yang, Peng Gao “Research on the Short-term Traffic Flow
Prediction Method Based on BP Neural Networks”2009.
[17] Klaus P. Hochheim, David G. Barber, Paul R. Bullock “Spring Wheat Yield Prediction for
Western Canada Using Weekly NOAA AVHRR Composites” 1996 IEEE.
[18] Laura Dente, Michele Rinaldi, Francesco Mattia, Giuseppe Satalino “Assimilation of ASAR
Data for Wheat Yield Prediction: Matera Case Study” 2005 IEEE.
[19] CAO Weixing, TANG Liang, ZHU Yan, PAN Jie, LI Weiguo, CHEN Binling
“Development of Growth Model-based Decision Support System for Crop Management” Second
International Symposium on Plant Growth Modeling, Simulation, Visualization and
Applications, 2007 IEEE.
[20] Jianqiang REN, Zhongxin CHEN, Xiaomei YANG, Xingren LIU, Qingbo ZHOU “Regional
Yield Prediction of Winter Wheat Based on Retrieval of Leaf Area Index by Remote sensing
Techonology” 2009 IEEE.
[21] TENG Fei, CHEN Zhongxin, WU Wenbin, Huang Qing, Li Dandan, Xia Tian “Yield
estimation of winter wheat in North China Plain by using crop growth monitoring system
(CGMS)” International S&T Cooperation Projects of China, the National Natural Science
Foundation of China, 2011.
[22] Ranjana Agarwal, Forcasting Techniques in Crops, Library Avenue, New Dalhi 2008.
[23] E. Houshyar, M.J.Sheikh Davoodi “Energy use Forcasting for Wheat Production utilizing
Artificial Neural Network” World Applied sciences Journal 10(8): 958-962,2010.
[24] Daniel Svozil, Vladimir KvasniEka, JiE Pospichal “Introduction to multi-layer feed-forward
neural networks” Tutorial, Elsevier 1997.
[25] Jeetendra Pande, Bhavika Verma, “Survey of Crop Yield Estimation Models with Emphasis
on Artificial Neural Network Model” 2nd International Conference, INDIAcom 2008.
[26] Georg Rus, Rudolf Kruse, Martin Schneider, and Peter Wagner “Estimation of Neural
Network Parameters for Wheat Yield Prediction” International Federation for Information
Processing, Volume 276, pp. 109–118 Springer 2008.
[27] George A. Hanuschak Sr.” Timely and Accurate Crop Yield Forecasting and Estimation
History and Initial Gap 2009” 2013 IEEE.