Similarity Data Item Set Approach: An Encoded Temporal Data Base Technique

95 Similarity Data Item Set Approach: An Encoded Temporal Data Base Technique
M.S.Danessh, C. Balasubramanian and K. Duraiswamy Abstract ‐ Data mining has been widely recognized as a powerful tool to explore added value from large-scale databases. Finding frequent item sets in databases is a crucial in data mining process of extracting association rules. Many algorithms were developed to find the frequent item sets. This paper presents a summary and a comparative study of the available FP-growth algorithm variations produced for mining frequent item sets showing their capabilities and efficiency in terms of time and memory consumption on association rule mining by taking application of specific information into account. It proposes pattern growth mining paradigm based FP-tree growth algorithm, which employs a tree structure to compress the database. The performance study shows that the anti- FP-growth method is efficient and scalable for mining both long and short frequent patterns and is about an order of magnitude faster than the Apriority algorithm and also faster than some recently reported new frequent-pattern mining.

Keywords:  Encoding method, frequent pattern mining, FP growth, FP tax, anti FP growth algorithm  ——————————  ——————————   1 INTRODUCTION

One of the currently fastest and most popular algorithms for frequent item set mining is the FP-growth algorithm. It is based on a prefix tree representation of the given database of transactions (called an FP-tree), which can save considerable amounts of memory for storing the transactions. The basic idea of the FP-growth algorithm can be described as a recursive elimination scheme in a preprocessing step delete all items from the transactions that are not frequent individually i.e., do not appear in a user-specified minimum number of transactions. Recourses to process the obtained reduced (also known as projected) database, remembering that the item sets found in the recursion share the deleted item as a prefix. On return, remove the processed item also from the database of all transactions and start over, i.e., process the second frequent item etc. In these processing steps the prefix tree, which is enhanced by links between the branches, is exploited to quickly find the transactions containing a given item and also to remove this item from the transactions after it has been processed[4][7].
              The  Apriori  heuristic  achieves  good  performance  gained by (possibly significantly) reducing the size of  candidate sets [3]. However, in situations with a large  number of frequent patterns, long patterns, or  quite  low  minimum  support  thresholds,  compact  data  structure,  called  frequent‐pattern  tree,  or  FP‐tree  in  short is constructed, which is an extended prefix‐tree  structure  storing  crucial,  quantitative  information  about  frequent  patterns.  To  ensure  that  the  tree  structure  is  compact  and  informative  only  frequent  length‐1  items  will  have  nodes  in  the  tree,  and  the  tree  nodes  are  arranged  in  such  a  way  that  more  frequently  occurring  nodes  will  have  better  chances  of node sharing than less frequently occurring ones.  This  experiments  show  that  such  a  tree  is  compact  and  it  sometimes  orders  of  magnitude  smaller  than  the original database [7]. Subsequent frequent‐pattern  mining will only need to work on the FP‐tree instead  of  the  whole  data  set.  The  properties  of  FP‐tree  are  thoroughly  studied  [10].  Also,  it  point  out  the  fact  that,  although  it  is  often  compact,  FP‐tree  may  not  always be minimal. Some optimizations are proposed  to speed up FP‐growth which is a technique to handle  single  path  FP‐tree  has  been  further  developed  for  performance improvements.     A  database  projection  method  has  been  developed  in  Section  2  to  cope  with  the  situation  when an FP‐tree cannot be held in main memory the  case that may happen in a very large database.    Extensive  experimental  results  have  been  reported.  Thus  examine  the  size  of  FP‐tree  as    Well  as  the  turning  point  of  FP‐growth  on  data  projection  to  building  FP‐tree[9].  The  main  step  is  described  in  Section 3, namely how an FP‐tree is projected in order 

1. PG student, Dept of CSE, K. S. R. College of Technology,Tiruchengode, Tamilnadu, India.
2. Asst.Professor,Dept of CSE, K.S.R. College of Technology, Tiruchengode, Tamilnadu, India. 3.Dean(academic),Dept of CSE, K.S.R. College of Technology,Tiruchengode, Tamilnadu, India.

JOURNAL OF COMPUTING, VOLUME 2, ISSUE 3, MARCH 2010, ISSN 2151-9617 HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/

96 to  obtain  an  FP‐tree  of  the  (sub)  database  containing  the  transactions  with  a  specific  item  (though  with 

…(Full text truncated)…