Frequent Item Set Mining with TM Algorithm and Tree Creation

Frequent Item Set Mining with TM Algorithm and Tree Creation

G. Ramesh KumarK. Arulanandam A. Kavitha

PG & Research, Dept of Computer Science & Applications, Govt Thirumagal Mills College, Thiruvalluvar University, Vellore, India

PG & Research Dept of Computer Science and Applications, Govt Thirumagal Mills College, Gudiyatham, India

PG &Research Department of Computer Science and Applications, DG Vaishnav College, Arumbakkam, Chennai-106, India

Corresponding Author Email: 
grk92804@rediffmail.com
Page: 
171-175
|
DOI: 
https://doi.org/10.18280/ama_b.610401
Received: 
10 September 2017
|
Accepted: 
16 March 2018
|
Published: 
31 December 2018
| Citation

OPEN ACCESS

Abstract: 

Item set mining is a skill widely recycled in data mining for determining cherished correlations amongst data. The useful measure to extract the knowledge based on user interest is by means of frequency. This is the mostly used technique in order to get the data based on the user preferences and user request, so I have proposed frequent item set mining for searching key element, here mining top-k frequent closed item sets without minimum support should be more preferable than the traditional minimum support-based mining. The recital and suppleness for mining top-k frequent closed item sets, as well as mining top-k frequent closed item sets in data stream milieus and mining top-k frequent closed chronological or tight patterns. Mining top-k numerous closed item sets of Length no less than k value it will indiscriminately quarried.

This paper introduces a novel calculation for mining complete continuous item sets. This calculation is alluded to as the TM (Transaction Mapping) calculation from here on. In this calculation, exchange ids of everything set are mapped and compacted to nonstop exchange interims in an alternate space and the numbering of item sets is performed by crossing these interim records in a profundity first request along the lexicographic tree. At the point, when the pressure coefficient winds up noticeably littler than the normal number of comparisons for interims crossing point at a specific level, the calculation changes to exchange and convergence.

The calculation against two prominent continuous things set mining calculations, FP-development, utilizing an assortment of informational indexes with short and long successive examples. Exploratory information demonstrates that the TM calculation outflanks these two calculations.

Keywords: 

Frequent Item, FP-Growth, Support and Confident

1. Introduction
2. Literature Survey
3. Proposed Work
4. Performance Analysis
5. Conclusion
  References

[1] Han J, Pei J, Yin Y. (2000). Mining frequent patterns without candidate generation. Proc. ACM SIGMOD Int’l Conf. Management of Data, pp. 1-12. 

[2] Bonchi F, Giannotti F, Mazzanti A, Pedreschi D. (2005). ExAnte: A preprocessing method for frequent-pattern mining. Intelligent System 20(3): 25–31. https://doi.org/10.1109/MIS.2005.45

[3] Bonchi F, Goethals B. (2004). FP-Bonsai: The art of growing and pruning small FP-trees. Proc. 8th Pacific-Asia Conf. Adv. Knowl. Discovery Data Mining, pp. 155–160.

[4] Agrawal R, Srikant R. (1994). Fast algorithms for mining association rules. Proc. 20th Int’l Conf. Very Large Data Bases (VLDB), pp. 487-499.

[5] Pei J, Han J, Mortazavi-Asl B, Wang JY, Pinto H, Chen QM, Dayal U, Hsu MC. (2004). Mining sequential patterns by pattern-growth: The prefix span approach. Knowledge and Data Eng 16(10): 424-1440. https://doi.org/10.1109/TKDE.2004.77

[6] Li PX, Chen JP, Bian FL. (2004). A developed algorithm of a priori based on association analysis. Geospatial Information Science 7(2): 108-112.

[7] Wang JY, Han JW, Lu Y, Tzvetkov P. (2005). TFP: An efficient algorithm for mining top-k frequent closed itemsets. Transactions on Knowledge and Data Engineering 17(5). https://doi.org/10.1109/tkde.2005.81

[8] Tseng VS. et al. (2013). Efficient algorithms for mining high utility item sets from transactional databases. Knowledge and Data Engineering 25(8).

[9] Cagliero L, Garza P. (2014). Infrequent weighted item set mining using frequent pattern growth. Knowledge and Data Engineering 26(4).

[10] Liu JQ, Wang K, Fung BCM. (2016). Mining high utility patterns in one phase without generating candidates. Knowledge and Data Engineering 28(5): 1245-1257. https://doi.org/10.1109/TKDE.2015.2510012

[11] Janarthanan P, Rajkumar N, Padmanaban G, Yamini S. (2014). Performance analysis on graph based information retrieval approaches. AMSE Journal –Series: Advances-D 19(1): 1-14.