臺灣博碩士論文加值系統

Pet animals play an important role in many people’s lives. Every single year, it occurred increasing enthusiasm of people to get pet. Recently, according to the American Pet Products Association (APPA), almost 85 million households have a pet. Obviously, trends have pets getting popular recently. Many people have pets to enhance relationships with partners, children or family members. Many elderly people like to have a pet as companionship because of loneliness. Adoption in animal shelter is one of the ways that common people do to get pet animal. The animal shelter provides various types of pets based on the adopter needed. The aim of this study was to know the behavior adopter pattern for pet adoption. This study only focused on cat and dog animals. Apriori association rule mining was used for finding interesting pattern from the animal shelter database. Moreover, this study generated models to predict how long pet animal stayed in shelter using machine learning approach from variables found to be important predictor in adoption. Intake reason, age, color, and primary breed were major determinants of adoption in cats. Intake reason, primary breed, color, secondary breed, and age were major determinants of adoption in dogs. Knowing the length of stay of animal in shelter was important, since in the animal shelter has space limitations. In case that it occurred overpopulation so the other problem would arise as maintenance and social problems. The results of this study could help animal shelter employees to increase level of participant of adopters. Once the behavior of adopter patterns was obtained, it could determine which characteristics of cat or dog have in demand by adopters. Furthermore, in case the length of stay of prediction result was long term, it could help animal shelter employees for making good decision whether to defend the animal in the shelter, or take other actions such as sale or donate the animal for health purpose.

ABSTRACT i
Dedication iii
Acknowledgments iv
CONTENTS v
List of Figures vii
List of Tables viii
Chapter 1 Introduction 1
1.1 Background 1
1.2 Previous Related Works 4
1.3 Research Objective 6
1.4 Limitations 6
1.5 Thesis Development 7
Chapter 2 Literature Study 8
2.1 Length of Stay 8
2.2 Statistical Analysis - Chi Square Test 8
2.3 Machine Learning Method 10
2.3.1 Decision Tree 10
2.3.2 Naïve Bayes 12
2.3.3 Support Vector Classifier (SVC) 13
2.4 Ensemble Learning – AdaBoost Algorithm 16
2.5 K-Fold Cross Validation 17
2.6 Apriori Algorithm 19
Chapter 3 Proposed Method 22
3.1 System Design 22
3.2 Data Preparation 23
3.3 Pre-processing Data 24
3.4 Feature Extraction 28
3.5 Classification Method 29
3.6 Performance Measurement 29
3.7 Data Mining Process 31
Chapter 4 Experimental Results 32
4.1 Chi Square Test Results 32
4.1.1 LOS and Colors 32
4.1.2 LOS and Intake Reason 33
4.1.3 LOS and Primary Breed 34
4.1.4 LOS and Secondary Breed 35
4.1.5 LOS and Age 36
4.1.6 LOS and Gender 37
4.1.7 LOS and Coat Length 38
4.1.8 LOS and Type of Dog 39
4.1.9 Analysis Chi Square Test Results 39
4.2 Information Gain 41
4.3 Predicting Results 43
4.4 Association Rules Results 45
Chapter 5 Conclusions and Future Work 51
5.1 Conclusions 51
5.2 Future Work 51
References 52
Appendix 55
Appendix 1 Cat Categorization 55
Appendix 2 Dog Categorization 64

[1] I. Dimitrijević, “Animal-assisted therapy–a new trend in the treatment of children and adults,” Psychiatria Danubina, vol. 21, no. 2, pp.236-241, Mar. 2009.
[2] J. Baxter and J. Rowe, “APPA releases findings from new 2019-2020 National Pet Owners Survey,” American Pet Prodcuts Association, Mar. 2019. [Online]. Available: https://americanpetproducts.org/press_releasedetail.asp?id=192. [Accessed June 22, 2019].
[3] R. Kestenbaum, “The biggest trends in the pet industry,” Forbes, Nov. 2018. [Online]. Available: https://www.forbes.com/sites/richardkestenbaum/2018/11/27/ the-biggest-trends-in-the-pet-industry/#63656f3ff099. [Accessed June 22, 2019].
[4] “Homelessness issues,” American Society for the Prevention of Cruelty to Animals, [Online]. Available: https://www.aspca.org/animal-homelessness. [Accessed June 22, 2019].
[5] “Pets by the numbers,” Animal Sheltering, [Online]. Available: https://www.animalsheltering.org/ page/pets-by-the-numbers [Accessed June 22, 2019].
[6] A. Protopopova, L. R. Mehrkam, M. M. Boggess and C. D. L. Wynne, “In-kennel behavior predicts length of stay in shelter dogs,” PloS One, vol. 12, no. 12, Dec. 2014.
[7] D. L. Wells, “A review of environmental enrichment for kennelled dogs, Canis familiaris,” Applied Animal Behaviour Science, vol. 85, no. 3-4, pp.307-317, Mar. 2004.
[8] R. Jennings, “Stray dogs hound taiwan after euthanasia ban takes effect,” The New Lens, Jun. 2018. [Online]. Available: https://international.thenewslens.com/ article/96625 [Accessed June 22, 2019].
[9] M. Lepper, P.H. Kass and L.A. Hart, “Prediction of adoption versus euthanasia among dogs and cats in a California animal shelter,” Journal of Applied Animal Welfare Science, vol. 5, no. 1, pp.29-42, Jun.2010.
[10] A. Protopopova, A. J. Gilmour, R. H. Weiss, J. Y. Shen and C. D. L. Wynne, “The effects of social training and other factors on adoption success of shelter dogs,” Applied Animal Behaviour Science, vol. 142, no. 1-2, pp.61-68, Dec. 2012.
[11] D. Wells, and P. G. Hepper, “The behaviour of dogs in a rescue shelter,” Animal Welfare, vol. 1, no. 3, pp.171-186, Aug. 1992.
[12] W. P. Brown, J. P. Davidson and M. E. Zuefle, “Effects of phenotypic characteristics on the length of stay of dogs at two no kill animal shelters,” Journal of Applied Animal Welfare Science, vol. 16, no. 1, pp.2-18, Jan. 2013.
[13] E. Weiss, K. Miller, H. Mohan-Gibbons and C. Vela, “Why did you choose this pet?: adopters and pet selection preferences in five animal shelters in the United States,” Animals, vol. 2, no. 2, pp.144-159, Apr. 2012.
[14] M. D. Salman, J. G. New, Jr, J. M. Scarlett, P. H. Kass, R. Ruch-Gallie and S. Hetts, “Human and animal factors related to relinquishment of dogs and cats in 12 selected animal shelters in the United States,” Journal of Applied Animal Welfare Science, vol. 1, no. 3, pp.207-226, Jun 2010.
[15] N. Gourkow and D. Fraser, “The effect of housing and handling practices on the welfare, behaviour and selection of domestic cats (Felis sylvestris catus) by adopters in an animal shelter,” Animal Welfare, vol. 15, pp.371-377, Nov. 2006.
[16] J. R. Quinlan, “Discovering rules by induction from large collections of examples,” Expert Systems in the Micro Electronics Age, 1979.
[17] J.R. Quinlan, “Induction of decision trees,” Machine Learning, vol. 1, pp.81–106, Mar. 1986.
[18] I. Rish, “An empirical study of the naïve bayes classifier,” in Proc. of Int. Joint Conf. on Artificial Intelligence, New York, United States, vol. 3, no. 22, pp.41-46, Aug. 2001.
[19] C. J. C. Burgers, “A tutorial on support vector machines for pattern recognition,” Data Mining and Knowledge Discover, vol. 2, no. 2, pp.121-167, 1998.
[20] G. James, D. Witten, T. Hastie and R. Tibshirani, “An introduction to statistical learning, with applications in R,” Springer, 2013.
[21] R. Agrawal and R. Srikant, “Mining sequential patterns,” in Proc. of the Eleventh International Conference on Data Engineering, Taipei, Taiwan, vol. 95, pp.3-14, Mar. 1995.
[22] “Missing Values in Data,” Statistics Solutions, [Online]. Available at: https://www.statisticssolutions.com/missing-values-in-data/. [Accessed June 18, 2019].
[23] G. Chandrashekar and F. Sahin, “A survey on feature selection methods,” Computers and Electrical Engineering, vol. 40, no. 1, pp.16-28, Jan. 2014.