American Journal of Sports Science
Volume 3, Issue 5, September 2015, Pages: 93-97

The Use of Modular Feed Forward Neural Networks in Anticipating the Results of Handball Championship 2015

Amr Hassan1, 2

1Department of Sports Training, Faculty of Sports Education, Mansoura University, Mansoura, Egypt

2Institute of Sport Science, University of Graz, Graz, Austria

Email address:

To cite this article:

Amr Hassan. The Use of Modular Feed Forward Neural Networks in Anticipating the Results of Handball Championship 2015.American Journal of Sports Science.Vol.3, No. 5, 2015, pp. 93-97. doi: 10.11648/j.ajss.20150305.13


Abstract: Observation is a highly recommended approach in game analysis as it helps form a better understanding for the types of relations within the game. The aim of this study is to present a new approach for predicting competitions results which are based on game analysis by the use of Modular Forward Neural Networks (MFNN). The data of 80 games were analyzed (i.e. Fast break, Breakthrough, different type of shot…). The Data used to train Modular Feed Forward networks include 21 processing elements (PEs) as input, one element as output, 2 hidden layers, 100 epochs – termination Cross Validation, random initial weights, and weight update batch. The MFNN test contains single output case threshold 0, 5 on level 1000. Results show significant correlation between game results and neural network output 0.93, 0.96. Actual network output was 0, 91. Normalized Root Mean Square Error was 0,078. Final mean squared error was 0.9. The variables mostly affecting the results of (MFNN) were: fast breaks, and blocked shots. Using MFNN in predicting game results based on game details is considered a novel approach for evaluating the level of teams and competitors and for improving the training plans and tactics.

Keywords: Team Handball, Neural Networks, Anticipation


1. Introduction

During the last decade, it seems that the handball society has significantly benefited from game analysis through the regular and disciplined match analysis along with team-success determining factors of techniques, tactics and workouts. Monitoring rivals teams and evaluating them by various numeric values help re-arrange workouts. The game analysis studies in handball were classified into: Free analysis, acoustic analysis, written analysis, film analysis, video analysis, video/computer analysis and computer analysis [1]. Few literatures are interested in figuring out the relations between game details, as well as ways of investing the huge output numeric data in predicting game context. Research in game analysis has occupied a unique place  in the field of sports science for a long time [2]; [3]; [4].

Neural Networks have been widely used in game analysis as they shed light on ways of using the data of game analysis effectively. Typical examples of this approach are tactical analysis and group behaviors [5]. Some literatures observed new approach of game analysis based on Custom Made Software (MASA) which is specific to determine players’ positions based on tactical analysis [6] ; [7].  Other literatures studied the interaction between offensive and defensive group tactics [8].

Performance and results predictions were not challenging for specialists and if this challenge took place, specialists did not resort to the use of modern methods such as advanced nonlinear modeling techniques [9] ; [10]. A substantial amount of exertion should be spent on foreseeing the results of sporting events [11]. Other researcher assumed that utilizing the neural systems helps rate and select particular players for specific rivals [12] ; [13].

From this point of view the previous literatures did not pay attention to observed methods of predicting games results based on previous analytical data from game context. Therefore the aim of this study is to give new methods for anticipating game championship results by means of Modular Feed Forward Neural Networks.

The Modular Feed Forward Neural Networks (MFNN)

MFNN is considered a special class of a multilayer perceptron (MLP), which processes the input by utilizing a few parallel MLPs, and recombines the outcomes. This has a tendency to make some structures inside of the topology cultivate specialization of capacity in every sub-module. As opposed to the MLP, particular systems don't have a full interconnectivity between their layers. Thus, a fewer number of weights are needed for the same size system (i.e. the same number of PEs). This has a tendency to accelerate preparing times and to diminish the quantity of obliged preparing models. There are numerous approaches to portion an MLP into the module. It is hazy how to plan the measured topology in light of the information. It cannot be certain that every module is practicing its preparation for a one of a kind part of the data [14];[15].

2. Material & Methods

Out of 160 half games from the handball world cup championship 2015, 149 half games were analyzed, the only half - games finished with a win or lose were considered, no equal games are included in this study. Eighteen quantity / quality game variables were determined as shown in table (1).

Table 1. Quantity / Quality game variables and desired results of half games for each team.

Input Data Desired Data
Team Total Shot 6M Goal 6M Shot Wing Shot Goal Wing Shot 9M Goal 9M Shot 7M Goal 7M Shot Fast Break Goal Fast Break shot Breakthroughs Goal Breakthroughs Shot Assists Technical Fault Steals Blocked Shots 2 Minute Suspensions Lose Win
ALG 331 29 43 26 44 42 156 18 26 32 45 15 17 87 89 36 12 20 9 1
ARG 266 30 50 26 35 34 90 17 26 26 36 19 29 75 79 30 11 20 5 6
AUT 296 24 41 25 43 41 96 16 20 48 69 20 27 92 78 31 16 20 6 4
BIH 320 51 77 16 35 44 116 16 26 30 40 18 26 103 102 32 17 21 8 2
BLR 355 72 114 27 40 34 104 18 24 37 43 27 30 103 99 28 15 22 7 3
BRA 307 40 73 17 29 35 109 15 19 38 48 26 29 86 88 33 6 18 8 4
CHI 321 45 75 8 22 33 122 25 36 25 32 29 34 74 117 32 11 20 7 3
CRO 410 38 56 37 55 78 174 29 39 56 64 20 22 116 109 39 42 26 5 13
CZE 348 22 38 22 40 55 130 27 34 62 76 21 30 111 89 30 30 22 6 4
DEN 413 73 102 32 38 80 169 23 27 46 54 18 23 152 92 23 29 28 2 12
EGY 271 24 47 17 32 49 111 18 27 17 22 26 32 63 63 25 16 18 6 4
ESP 419 61 78 40 54 60 155 31 41 49 60 24 31 151 119 38 42 26 3 13
FRA 413 53 81 34 48 50 132 40 48 44 58 38 46 131 94 52 32 26 3 14
GER 386 58 86 28 39 45 116 28 38 56 67 35 40 138 105 29 35 30 6 10
IRI 358 36 53 17 29 65 186 17 25 37 49 13 16 91 115 30 17 16 8 1
ISL 296 28 45 24 39 49 139 14 18 27 36 10 19 93 52 22 12 18 7 4
KSA 333 24 46 10 32 40 170 12 21 30 38 20 26 51 118 30 8 21 10 0
MKD 284 40 53 29 39 42 103 23 31 31 41 16 17 109 65 16 10 18 4 7
POL 404 66 92 21 30 68 170 17 26 43 53 26 33 112 117 48 31 30 8 9
QAT 426 75 113 10 20 75 195 26 33 22 25 37 40 114 105 38 9 25 8 10
RUS 339 33 53 31 51 54 128 23 29 43 50 21 28 97 86 31 18 17 5 5
SLO 417 50 77 36 54 35 99 32 40 64 81 56 66 149 122 39 33 29 10 8
SWE 261 31 38 14 29 44 107 19 23 36 43 13 21 102 80 29 22 18 2 8
TUN 292 40 64 19 32 43 123 11 20 23 31 16 22 69 54 21 5 19 6 4

Network Design, Training and Test

The network design and training procedures contain 21 PEs as input, one as output, sixteen exemplars, two hidden layers, 100 Epochs Termination Cross Validation, random initial weights and weight update batch. The MFNN test contains single output case threshold 0.5 on level 1000 Epochs (as shown in fig 1).

Figure 1. Graphical user interface of the (MFNN).

The outputs of championship analysis (eighteen quantity / quality variables) were used to feed the neural network as input data, also game results (win / lose) were marked as desired data. Up to 70% of all data were marked as training, while 15% were marked as cross validation. Cross validation processes the lapse in a test set while the system is being prepared with the preparation set. It is realized that the mean squared normalized error (MSE) will continue to diminish in the preparation set, however, it may begin to increase in the test set. This happens when the system begins "remembering" the preparation designs. The Termination page of the enactment control reviewer can be utilized to screen the cross approval set mistake and consequently it stops the system when it is not progressing. 15 % of all data were marked as testing data. The sufficient correlation between desired and output results from the neural network was calculated. The network train report also contains minimum training MSE at last epoch (table. 2).

Table 2. Mean squared normalized error for training and cross validation.

Best Networks Training Cross Validation
Epoch # 8 1
Minimum MSE 5.74129E-29 0.022651597
Final MSE 5.74129E-29 0.053782498

3. Results

There is no significant different between the mean of actual championship results and the MFNN results. P value was 0.005 in case of win or lose as show in table 3.

Table 3. The mean and standard deviation of actual and (MFNN) output data, P values (2-tailed test) also shown.

Half games Actual data Network output P
  Mean SD Mean SD  
Win 6,2 1.1 5.58 0.97 0.005
Lose 6,2 1.1 5.76 0.99 0.005

Test report on network performance shows values of Root Mean Square Error (RMSE), it was between 2.5 in case of win and 2.3 in case of lose.  Normalized Root Mean Square Error (NRMSE) and Mean Absolute Error (MAE) are also shown.  Correlation between the actual results and the MFNN results was very high. In case of win, it was 0.930 and in case of lose, it was 0.960 as shown in table 4.

Table 4. The test report error of the neural network shows the sufficient correlation between desired competition results (win / lose in each half game) and (MFNN) output data.

Performance Win Lose
RMSE 2.517 2.316
NRMSE 0.209 0.217
MAE 2.272 2.649
Min ABS Error 0.180 0.183
Max ABS Error 3.382 3.303
R 0.930 0.960

The accuracy of actual competition results and the MFNN output are shown in fig 2 and the game data affecting the MFNN accuracy are shown in fig 3.

Figure 2. Championship game results and actual (MFNN) output. (I.e. Win for each half game)

Figure 3. The Data of Half Game (Lose) Affecting (MFNN) output.

4. Discussion

The previous presentation of the results shows the reliability of the MFNN methodology in predicting the results of games in the competition. Previous studies relied on the prediction of athletic performance only without the possibility of pre-research to identify the results of matches, therefore, they used a hybrid prediction system based on genetic algorithm and artificial neural network (GANN) [12], or the neural networks to select players. Both are based on tests in advance [13].

This study tried to anticipate the results of the handball world cup championship 2015 by using modular feed forward neural networks and relied on neural network feed variables resulting from the analysis of the tournament and number 18 variables from game events, The results show that there are no significant differences between the expected output of the MFNN and the actual data of the competitions. Results also showed the presence of a high correlation between the output results by using MFNN and actual results for each half game in all cases of winning and losing as 93% and 96%, respectively.

Also, the mean square error of the neural network was at its lowest value 1.8, which increased the confidence in the results presented. In the same context, the sensitivity of MFNN input data to game (win/lose) will help trainers and experts to understand the nature of the biggest factors influencing the outcome of matches and thus they work to take advantage of playing through the development of training programs and plans. Another point of view adopts the possibility of the use of MFNN to evaluate team performance in a phased manner during the competitions. It will identify the level at which the team is heading and its competitors. This will help us make quick decisions that will change the course of the results and the team.

The prospective values of this study is to show that the approach of artificial Feed Forward Neural network dealing with the methods of optimal use of the data in the field of sports based on performance analysis is a reliable approach in predicting game results. The performance output data are necessary as a base to predict other values in cooperation with MFNN. Using this proposed methodology of MFNN helps predict large values which in turn will lead us to improve the player’s performance, (I. e.) Type of tactics.

5. Conclusion and Future Work

The Modular Forward Neural Network was used for game execution forecast. The test results can help the trainers to nearly foresee the performance of the team and the competitors. Any future research work ought to look at some genuine information from distinctive games and diverse levels. Moreover, the researchers ought to enhance the calculation to make the expectation more exact. At long last, it ought to look at different methodologies for expectations.


References

  1. Y. Taskiran. (2007). "2007 EHF Youth Coaches´ Course during the 2007 W 19 European Championship," Coach. E H F Youth Championship, Eur.
  2. A. Hohmann and M. Lames. (2005). "Trainings-und Wettspielanalyse," M. K. & K. R. In A. Hohmann and (Hrsg.), Eds.Handbuch Sportspiel. - Schorndorf : Hofmann, pp. 376–394.
  3. D. Memmert and K. Roth.(2003). "Individualtaktische Leistungsdiagnostik im Sportspiel," Spectrum, vol. 15, no.July, pp. 44–70.
  4. W. Schöllhorn and P.  Jürgen. (2002). "Prozessanalysen in der Bewegungs- und Sportspielforschung —Sportinformatik," Spectr. der Sport., vol. 14, no. 1, pp. 30–52.
  5. M. Pfeiffer and J. Perl. (2006). "Analysis of tactical Structures in team handball by means of artificial neural networks," Int. J. Comput. Sci. Sport, vol. 5, no. 1, pp. 4–14.
  6. P. Rudelsdorfer, N. Schrapf, H. Possegger, T. Mauthner, H. Bischof, and M. Tilp.(2014). "A novel method for the analysis of sequential actions in team handball," Int. J. Comp. Sci. Sport, vol. 13, no. 1, pp. 69–84.
  7. N, Scharpf., M, Tilp. (2013). "Action sequence analysis in team handball," J. Hum. Sport Exerc. North Am., vol. 8, no. 3Proc, pp. 615–621.
  8. S, Alsaied., N, Scharpf., A, Hassan., M, Tilp. (2015). "Analysis Of Interaction Between Offense And Defence Tactics In Team Handball By Means Of Artificial Neural Networks," in 20th Annual ECSS-Congress, Malmö.
  9. M. Janta, C. Ebert, and V. Senner. (2012). "Functionality and performance of customized sole inlays for various sports applications," Procedia Eng., vol. 34, pp. 290–294.
  10. Y. Torun and G. Tas. (2012). "Remote Sensing Image Classification By Non-Parallel Svms," Adv. Inf. Technol. andManagement, vol. 1, no. 3, pp. 1–3.
  11. H. O. Stekler, D. Sendor, and R. Verlander.(2010). "Issues in sports forecasting," Int. J. Forecast., vol. 26, no. 3, pp. 606–621.
  12. X. Biao. (2012). "Prediction of Sports Performance based on Genetic Algorithm and Artificial Neural Network," Int. J. Digit. Content Technol. its Appl., vol. 6, no. 22, pp. 141–149.
  13. S. R. Iyer and R. Sharda. (2009). "Prediction of athletes performance using neural networks: An application in cricket team selection," Expert Syst. Appl., vol. 36, no. 3, pp. 5510–5522.
  14. V. G. Ivancevic and T. T. Ivancevic. (2010). Quantum Neural Computation. Springer Science & Business Media,.
  15. A. O. Ali, I. A. Saleh, and T. R. Badawy. (2010). "Intelligent Adaptive Intrusion Detection Systems Using Neural Networks ( Comparitive study )," Int. J. Video& Image Process. Netw. Secur. IJVIPNS-IJENS, vol. 10, no. 01, pp. 1–8.

Article Tools
  Abstract
  PDF(809K)
Follow on us
ADDRESS
Science Publishing Group
548 FASHION AVENUE
NEW YORK, NY 10018
U.S.A.
Tel: (001)347-688-8931