Discovering the relationships between yarn and fabric properties using association rule mining

Investigation of the effects of yarn parameters on fabric quality and nding important parameters to achieve desired fabric properties are important issues for the design process with the aim to meet the needs of the textile industry and the consumer for complex and speci c requirements of functionality. Despite many statistical and mathematical studies that predict and reveal speci c properties of utilized yarn and fabric materials, a number of challenges continue to exist when evaluated in many perspectives, such as discovering complex relationships among material properties in data. Data mining plays an important role in discovering hidden patterns from fabric data and transforming it into knowledge. Therefore, the aim of the study is to uncover relationships between yarn parameters and fabric properties using an extended FP-Growth algorithm in association rule mining. This study extracts different types of frequent itemsets (closed, maximal, top-k, top-k closed, top-k maximal) that have not been determined in textile sector before. This article also proposes two novel concepts, closed frequent item and maximal frequent item, to identify signi cant items in data. In the experimental studies, the algorithm was executed on a real-world textile dataset with different support threshold values to compare the different types of patterns. Experimental results show that proposed approach is very useful for discovering rules related to yarn and fabric properties.

PDF

___

1] Chattopadhyay R. Arti cial neural networks in yarn property modeling. In: Majumdar A, editor. Soft Computing in Textile Engineering. Delhi, India: Woodhead Publishing, 2011. pp. 105-125.
[2] Shen J, Zhou X, Ma H, Chen W. Spectrophotometric prediction of pre-colored ber blends with a hybrid model based on arti cial neural network and Stearns-Noechel model. Text Res J 2017; 87: 296-304.
[3] Lee CKH, Choy KL, Ho GTS, Chin KS, Law KMY, Tse YK. A hybrid OLAP-association rule mining based quality management system for extracting defect patterns in the garment industry. Expert Syst Appl 2013; 40: 2435-2446.
[4] Huang Y, Qiu X, Yang J. The association rule mining technology in marketing decision-making and its application in logistics enterprises. In: Proceedings of the 4th International ISKE Conference on Intelligent Systems and Knowledge Engineering; 27{28 November 2009; Hasselt, Belgium: ISKE. pp. 252-257.
[5] Ingle MG, Suryavanshi NY. Association rule mining using improved Apriori algorithm. Int J Comput Appl 2015; 112: 37-42.
[6] Logeswari T, Valarmathi N, Sangeetha A, Masilamani M. Analysis of traditional and enhanced Apriori algorithms in association rule mining. Int J Comput Appl 2014; 87: 4-8.
[7] Ciarapcia FE, Sanctis ID, Resta B, Dotti S, Gaiardelli P, Bandinelli R, Fani V, Rinaldi R. Integrating sustainability in the fashion system using association rules. In: Rinaldi R, Bandinelli R, editors. Business Models and ICT Technologies for the Fashion Supply Chain, Lect Notes Electr En 2017; 413: 239-250.
[8] Cai Y, Cui X, Rodgers J, Thibodeaux D, Martin V, Watson M, Pang SS. A comparative study of the effects of cotton ber length parameters on modeling yarn properties. Text Res J 2013; 83: 961-970.
[9] Xue Z, Zeng X, Koehl L, Chen Y. Extracting fabric hand information from visual representations of ared skirts. Text Res J 2013; 84: 246-266.
[10] Yang S, Gordon S. Accurate prediction of cotton ring-spun yarn quality from high-volume instrument and mill processing data. Text Res J 2016; 1: 1-15.
[11] Huang XX, Tao XM, Xu BG. A theoretical model of maximum hairiness of staple ring-spun yarns. Text Res J 2014; 84: 1121-1137.
[12] Park S, Yun C, Kim J, Park CH. The effects of the fabric properties on fabric movement and the prediction of the fabric movements in a front-loading washer. Text Res J 2013; 83: 1201-1212.
[13] Kumar TS, Sampath V. Prediction of dimensional properties of weft knitted cardigan fabric by arti cial neural network system. J Ind Text 2012; 42: 446-458.
[14] Saharkhiz S, Abdorazaghi M. The performance of different clustering methods in the objective assessment of fabric pilling. J Eng Fiber Fabr 2012; 7: 35-41.
[15] Chenganmal M, Vijayakumar HL, Subramaniam V. Evaluation of bending rigidity of dressing materials using arti cial neural networks. Journal of Textile and Apparel Technology and Management 2013; 8: 1-10.
[16] Rahnama M, Semnani D, Zarrebini M. Measurement of the moisture and heat transfer rate in light-weight nonwoven fabrics using an intelligent model. Fibres Text East Eur 2013; 21: 89-94.
[17] Sun J, Yao M, Xu B, Bel P. Fabric wrinkle characterization and classi cation using modi ed wavelet coefficients and support-vector-machine classi ers. Text Res J 2011; 81: 902-913.
[18] Zhang J, Xungai W. Objective pilling evaluation of wool fabrics. Text Res J 2007; 77: 929-936.
[19] Agarwal G, Koehl L, Perwuelz A, Lee KS. Interaction of textile parameters, wash-ageing and fabric softener with mechanical properties of knitted fabrics and correlation with textile-hand - Interaction of textile parameters with laundry process. Fiber Polym 2011; 12: 670-678.
[20] Yildiz K, Buldu A, Demetgul M. A thermal-based defect classi cation method in textile fabrics with k-nearest neighbor algorithm. J Ind Text 2016; 45: 1-16.
[21] Bahadir SK, Kalaoglu F, Jevsnik S, Eryuruk SH, Saricam C. Use of arti cial neural networks for modelling the drape behaviour of woollen fabrics treated with dry nishing processes. Fibres Text East Eur 2015; 23: 90-99.
[22] Jing J, Zhang Z, Kang X, Jia J. Objective evaluation of fabric pilling based on wavelet transform and the local binary pattern. Text Res J 2012; 82: 1880-1887.
[23] Zhang J, Xin B, Fang H, Cao Y. Novel colour clustering method for interlaced multi-colored dyed yarn woven fabrics. Fibres Text East Eur 2015; 23: 107-114.
[24] Xiao Z, Nie X, Zhang F, Geng L, Wu J, Li Y. Automatic recognition for striped woven fabric pattern. J Text I 2014; 106: 409-416.
[25] Nourani G, Jeddi AAA, Moghadam MB. Determining the structural parameters and yarn type affecting tensile strength and abrasion of weft knitted fabrics using cluster analysis. Middle East J 2011; 8: 1008-1017.
[26] Agrawal R, Imielinski T, Swami A. Mining association rules between sets of items in large databases. In: Proceedings of the 1993 ACM SIGMOD International Conference on Management of data; 25{28 May 1993; Washington, DC, USA: ACM. pp. 207-216.
[27] Aggarwal CC, Bhuiyan MA, Hasan MA. Frequent pattern mining algorithms: A survey. In: Aggarwal CC, Han J, editors. Frequent Pattern Mining. Basel, Switzerland: Springer International Publishing, 2014. pp. 19-64.
[28] Zaki MJ, Hsiao CJ. CHARM: An efficient algorithm for closed itemset mining. In: Proceedings of the Second SIAM International Conference on Data Mining; 11{13 April 2002; Virginia, USA: SIAM. pp. 457-473.
[29] Burdick D, Calimlim M, Gehrke J. MAFIA: a maximal frequent itemset algorithm for transactional databases. In: Proceedings of the 17th International Conference on Data Engineering; 2{6 April 2011; Heidelberg, Germany: IEEE. pp. 443-452.
[30] Pietracaprina A, Riondato M, Upfal E, Vandin F. Mining top-K frequent itemsets through progressive sampling. Data Min Knowl Disc 2010; 21: 310-326.
[31] Cheung YL, Fu AWC. Mining frequent itemsets without support threshold: with and without item constraints. IEEE Trans Knowl Data Eng 2004; 16: 1052-1069.
[32] Li H, Wang Y, Zhang D, Zhang M, Chang E. PFP: Parallel FP-Growth for query recommendation. In: Proceedings of the ACM Conference on Recommender Systems; 23{25 October 2008; Lausanne, Switzerland: ACM. pp. 107-114.
[33] Yasar T. Hava jetli iplik egirme sistemleri ile uretilmis ipliklerin performanslarnn degerlendirilmesi. MSc, Erciyes University, Kayseri, Turkey, 2015.
[34] Liu H, Hussain F, Tan CL, Dash M. Discretization: An enabling technique. Data Min Knowl Disc 2002; 6: 393-423.