Association Rule Mining on Likert’s Scale Data using a Novel Attributes Pruning Technique

Ogedengbe M.T.¹ , Junaidu S.B.² , Kana A.F.D.³

1. Dept. of Computer Science, Joseph Sarwuan Tarka University, Makurdi, Nigeria.
2. Dept. of Computer Science, Ahmadu Bello University, Zaria, Nigeria.
3. Dept. of Computer Science, Ahmadu Bello University, Zaria, Nigeria.

Section:Research Paper, Product Type: Journal-Paper
Vol.12 , Issue.4 , pp.54-65, Aug-2024

Online published on Aug 31, 2024

Copyright © Ogedengbe M.T., Junaidu S.B., Kana A.F.D. . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
 

View this paper at   Google Scholar | DPI Digital Library

XML View     PDF Download

How to Cite this Paper

123 Views    104 Downloads    20 Downloads

Abstract :
In recent decades, the Apriori algorithm has emerged as a powerful tool for generating meaningful insights and supporting effective decision-making in data science. Traditionally a binary mining tool, Apriori is highly efficient in analysing transaction datasets in market basket analysis to uncover customer purchasing patterns. When applied to Likert scale datasets, however, it requires discretizing item attributes into binary values, which can result in significant information loss. This study proposes a novel mining technique called Common Question Attribute Pruning (CQAP), which enhances the standard Apriori algorithm by extending its capabilities to process 5-point Likert scale datasets without the need for attribute discretization, thereby preserving the ordinal nature of the respondents` opinions. The key innovation of this technique lies in its ability to represent all five points of the Likert scale within the Apriori framework, without converting them to Boolean transaction data (0s and 1s). The modified Apriori algorithm, termed the Extended Apriori Algorithm (Ext-AA), generates candidate sets by treating question-value (q,v) pairs as data points. During the candidate joining phase, any combinations with common question attributes are pruned. This approach introduces a new perspective on defining support count, minimum support threshold metric, and minimum confidence threshold metric, which helps in filtering out infrequent candidate sets and the determination of the strength of the association rules derived from the sample dataset. In experimental evaluations on sample datasets, the Ext-AA produced 8 strong rules, whereas the standard Apriori algorithm generated 135 rules which is 89.63% rule reduction after pruning. These results demonstrate the superior performance potential of CQAP technique against the state of art Apriori algorithm on the evaluated sample datasets.

Key-Words / Index Term :
Apriori; Pruning; Likert scale; Questionnaire; Candidate-set; Itemset; Support; Confidence

References :
[1] K. Baffour, C. Osei-Bonsu, A. Adekoya, “A Modified Apriori Algorithm for Fast and Accurate Generation of Frequent Item Sets”. International Journal of Scientific & Technology Research (IJSTR), Vol.6, pp.169-173, 2017.
[2] R. Z. Niu, G. Fang, D. Zeng, H. Yuan, “An improved Apriori algorithm based on influence weight”. In?Third International Conference on Computer Vision and Pattern Analysis (ICCPA 2023), Vol.12754, pp.673 679, 2023.SPIE.
[3] X. Wu, Y. Xiao, A. Liu, “Application of improved Apriori Algorithm in Innovation and Entrepreneurship Engineering Education Platform”.?Scalable Computing: Practice and Experience,?Vol.24, No.3, pp.609-620, 2023.
[4] E.S. Harpe, D. Pharm, “How to analyze Likert and other rating scale data”. Curr. Pharm. Teach. Learn. Vol.7, pp.836-850, 2015.
[5] U. Sivarajah, M. Kamal, Z. Irani, V. Weerakkody, “Critical Analysis of Big Data challenges and Analytical Methods”. Journal of Business Research (JBR), Published by Elsevier Inc. Vol.70, pp.263-286, 2017, DOI: https://doi.org/10.1016/j.jbusres.2016.08.001.
[6] S. Rana, M.N.I. Mondal, “A Seasonal and Multilevel Association Based Approach for Market Basket Analysis in Retail Supermarket”.?European Journal of Information Technologies and Computer Science,?Vol.1, No.4, pp.9-15, 2021.
[7] R. Agrawal, G. Johannes, G. Dimitrios, R. Prabhakar, “Automatic Subspace Clustering of High Dimensional Data for Data Mining Applications”. In SIGMOD ’98: Proceedings of the 1998 ACM SIGMOD. International Conference on Management of data. New York, NY, USA. ACM Press, pp.94-105, 1998.
[8] A. Joshi, S. Kale, S. Chandel, D.K. Pal, “Likert scale: Evaluating the Interdependent Effect for Likert Scale Items”, Appl. Sci. Technol, Vol.7, pp.396-403, 2015.
[9] K. Teymoornejad, “Improving Apriori Algorithm with Various Techniques”,2018. https://www.researchgate.net/publication/329736159_Improving_Apriori_Algorithm_with_various _techniques.
[10] R. Agrawal, T. Imieli?ski, A. Swami, “Mining association rules between sets of items in large databases”. Paper presented at the Proceedings of the 1993 ACM SIGMOD. International Conference on Management of data. 1993.
[11] B.P. Subedi, “Using Likert type data in social science research: confusion, issues and challenges”. Int.J. Contemp. Appl. Sci. Vol.3, pp.36-49, 2016.
[12] K. Singh, J. Maiti, “Mining Frequent Patterns with Temporal Effect: A Case of Accident Path Analysis”. Proceedings of ICETIT. Springer, pp.596-603, 2020.
[13] Z. Abdullah, A. Gusman, T. Herawan, M. Deris, “MILAR: Mining Indirect Least Association Rule Algorithm”. International Conference on Advanced Data and Information Engineering (ADIE), pp.159-166, 2016.
[14] C.H. Chee, J. Jaafar, I.A. Aziz, M.H. Hasan, W. Yeoh, “Algorithms for frequent itemset mining: a literature review”. Artificial Intelligence Review, Vol.4, pp.2603-2621, 2019.
[15] A.R. Riszky, M. Sadikin, “Data Mining using Apriori Algorithm for Product Recommendation for Customers”, Journal of computer system technology (JCST), Vol.7, pp.103-108, 2019. https://doi.org/ jtsiskom.7.3.2019.103-108.
[16] D. Patill, R. Deshmukh, D. Kirange, “Adaptive Apriori Algorithm for Frequent Itemset Mining”. Proceedings of the SMART, IEEE Conference ID: 39669 5th, International Conference on System Modeling & Advancement in Research Trends, College of Computing Sciences & Information Technology, Teerthanker Mahaveer University, Moradabad, India, pp.7-13, 2016.
[17] B. Yu, C.F. Wen, H.L. Lo, H.H. Liao, P.C. Wang, “Improvements in patient safety culture: a national Taiwanese survey”, International Journal for Quality in Health Care (IJQHC). Vol.16, 2020.
[18] D. Ginting, H. Mawengkang, S. Efendi, “Modification of Apriori Algorithm focused on Confidence Value to Association Rules”. Nommensen International Conference on Technology and Engineering. IOP Publishing IOP Conf. Series: Materials Science and Engineering, doi:10.1088/1757-899X/420/1/012125, 2018.
[19] A. Hodijah, U.T. Setijohatmo, G. Munawar, G.S. Ramadhanty, “Multi-threaded approach in generating frequent itemset of Apriori algorithm based on trie data structure”.?TELKOMNIKA (Telecommunication Computing Electronics and Control),?Vol.20, No.5, pp.1034-1045, 2022.
[20] R. Likert, “A technique for the measurement of attitudes”, Arch. Psychol, Vol.22, pp.55, 1932.
[21] A. Pandey, A. Jain, “Comparative analysis of KNN algorithm using various normalization techniques”, Int.J.Comp. Netw. Inf. Secur, Vol.11, pp.36-42, 2017. DOI:10.5815/ijcnis.2017.11.04.
[22] E. Hikmawati, N.U. Maulidevi, K. Surendro, “Pruning Strategy on Adaptive Rule Model by Sorting Utility Items”.?IEEE Access,?Vol.10, pp.91650-91662, 2022.
[23] M.T. Ogedengbe, S.B. Junaidu F.D.A. Kana, “Itemset Fusion: A New Technique for Itemsets Generation”. ATBU Journal of Science, Technology and Education, Vol.12, No.1, pp.216-230, 2024.
[24] D. Kriksciuniene, V. Sakalauskas, “AHP Model for Quality Evaluation of Healthcare System”. In: Damaevi?ius, R., Mikayt, V. (eds.) ICIST, Vol.756, pp.129-141. Springer, Cham. 2017. DOI:https://doi.org/10.1007/978-3-319-67642-511.
[25] R. Lewandowski, “Evaluating the Interdependent Effect for Likert Scale Items”. Paper presented at the Business Information Systems Workshops: BIS 2019 International Workshops, Seville, Spain, pp.26–28, 2020.
[26] P. Jalpa, D. Rustom, “A Novel Hybrid Method for Generating Association Rules for Stock Market Data”, 2017.
[27] Z. Kohzadi, A.M. Nickfarjam, L.S. Arani, M. Mahdia, “Extraction frequent patterns in trauma dataset based on automatic generation of minimum support and feature weighting”. BMC medical research methodology, Vol.24, No.1, pp.40, 2024.
[28] G.M. Sullivan, A.R. Artino Jr, “Analyzing and interpreting data from Likert-type scales”. Journal of graduate medical education (JGME), Vol.5, No.4, pp.541-542, 2013.
[29] K. Sumiran, “An Overview of Data Mining Techniques and their Application in Industrial Engineering”. Asian Journal. Appl. Sci. Technol, Vol.2, pp.947-953.
[30] R. Yadav, K Garg, “An Improved Multiple-level Association Rule Mining Algorithm with Boolean Transposed Database”. International Journal of Computer Science and Information Security (IJCSIS), Vol.13, pp.9, 2015.
[31] F. Li, C. Meng, C. Wang, S. Fan, “Equipment Quality Information Mining Method Based on Improved Apriori Algorithm”.?Journal of Sensors, 2023.
[32] M. Hosseini, O. Akbari, A. Rahmani, A. Shakeri, “Predicting Students’ Score in the English Section of the Ba Entrance Exam by Apriori Algorithm”. Research Journal of English Language and Literature (RJELAL), Vol.3, pp.306-313, 2015.
[33] N. Mahajan, J. Namdeo, “Analysis of Student Result Data Using Association Rule Mining”. Research Journal of Science and IT Management, (RJSITM), Vol.4, No.6, pp.32-37, 2015.
[34] C.L. Mao, S.L. Zou, J.H. Yin, “Educational Evaluation Based on Apriori-Gen Algorithm”. EURASIA Journal of Mathematics Science and Technology Education, Vol.13, No.10, pp.6555-6564, 2017. DOI: 10.12973/ejmste /78097.
[35] M.E. Saad, S.N.S. Abdullah M.Z. Murah, “Cyber Romance Scam Victimization Analysis using Routine Activity Theory Versus Apriori Algorithm”, International Journal of Advanced Computer Science and Applications (IJACSA), Vol.9, No.12, pp.479 – 485, 2018.
[36] K. Kakde, P. Gupta, “Performance Evaluation of Student Based on Attendance Using Data Mining Technique”, International Research Journal of Engineering and Technology (IRJET), Vol.5, No.5, pp.465-467, 2018.
[37] N.H. Hanafi, D.J. Juanis, M.H. Samian, S. Mohmad, S.A.M. Shafie, S.N. Shamsuddin, “An Application of Association Rule Mining in Analyzing Courses Affecting the Results of Financial Mathematics Case: UITM Seremban Campus”. Social Sciences, vol.9, pp.123-133, 2019.
[38] F. Asur, D. Sevimli, K. Yazici, “Visual Preferences Assessment of Landscape Character Types Using Data Mining Methods (Apriori Algorithm): The Case of Alt?nsaç and Inkoy (Van/Turkey)”. Journal of Agricultural Science and Technology (JAST), Vol.22, pp.247-260, 2020.
[39] D. Bertram, “Likert Scales”. http://my.ilstu.edu/~eostewa/497/Likert%20topic-dane-likert.pdf. 2016.
[40] A.P.U. Siahaan, A. Ikhwan, S. Aryza, “A Novelty of Data Mining for Promoting Education based on FP-Growth Algorithm”, 2018.
[41] S. Jain, S. Shukla, R. Wadhvani, “Dynamic selection of Normalization Techniques using Data Complexity Measures”, Expert System Application, Vol.106, pp.252-262, 2018.
[42] V.G. Manjunatha Guru, V. N. Kamalesh, K. B. Apoorva. "An Ensemble Learning Based Approach for Real-Time Face Mask Detection." Int. J. Sci. Res. in Computer Science and Engineering (IJSRCE), Vol.12, No.3, 2024.
[43] S. Pawan Wasnik, S.D. Khamitkar, Parag Bhalchandra, S.N. Lokhande, S. Ajit Adte, "An Observation of Different Algorithmic Technique of Association Rule and Clustering", International Journal of Scientific Research in Computer Science and Engineering (IJSRCSE), Vol.6, Issue.1, pp.28-30, 2018.
[44] M.T. Ogedengbe, S.B. Junaidu, A.F.D. Kana, "Adaptive Minimum Support Threshold for Association Rule Mining." Indonesian Journal of Data and Science (IJDS), Vol.5, No.2, 2024.
[45] S.A. Olaleye, D.C. Ukpabi, O. Olawumi, D.D. Atsa`am, R.O. Agjei, S.S. Oyelere, E.A. Kolog, Association rule mining for job seekers` profiles based on personality traits and Facebook usage International Journal of Business Information Systems (IJBIS), Vol.40, No.3, pp.299-326, 2022.
[46] L.M. Mohammed, M.T. Ogedengbe, “FP-Growth Algorithm: Mining Association Rules without Candidate Sets Generation.”, Kasu Journal of Computer Science, (KJCS), Vol.1, No.2, pp. 392-411.