An Efficient Frequent Patterns Mining Algorithm over Data Streams Based on FPD-Graph

2012 ◽  
Vol 433-440 ◽  
pp. 4457-4462 ◽  
Author(s):  
Jun Shan Tan ◽  
Zhu Fang Kuang ◽  
Guo Gui Yang
Keyword(s):  
Data Streams ◽  
Data Stream ◽  
Pattern Mining ◽  
Frequent Pattern Mining ◽  
Frequent Pattern ◽  
Frequent Patterns ◽  
Data Generation ◽  
Experiment Data ◽  
Mining Algorithm ◽  
Head Node

The design of synopses structure is an important issue of frequent patterns mining over data stream. A data stream synopses structure FPD-Graph which is based on directed graph is proposed in this paper. The FPD-Graph contains list head node FPDG-Head and list node FPDG-Node. The operations of FPD-Graph consist of insert operation and deletion operation. A frequent pattern mining algorithm DGFPM based on sliding window over data stream is proposed in this paper. The IBM synthesizes data generation which output customers shopping a data are adopted as experiment data. The DGFPM algorithm not only has high precision for mining frequent patterns, but also has low processing time.

scholarly journals Frequent Pattern Mining Algorithms for Finding Associated Frequent Patterns for Data Streams: A Survey

2014 ◽  
Vol 37 ◽  
pp. 109-116 ◽  
Author(s):  
Shamila Nasreen ◽  
Muhammad Awais Azam ◽  
Khurram Shehzad ◽  
Usman Naeem ◽  
Mustansar Ali Ghazanfar
Keyword(s):  
Data Streams ◽  
Pattern Mining ◽  
Frequent Pattern Mining ◽  
Frequent Pattern ◽  
Frequent Patterns ◽  
Mining Algorithms

scholarly journals WMFP-Outlier: An Efficient Maximal Frequent-Pattern-Based Outlier Detection Approach for Weighted Data Streams

2019 ◽  
Vol 48 (4) ◽  
pp. 505-521 ◽  
Author(s):  
Saihua Cai ◽  
Qian Li ◽  
Sicong Li ◽  
Gang Yuan ◽  
Ruizhi Sun
Keyword(s):  
Outlier Detection ◽  
Data Streams ◽  
Data Stream ◽  
Pattern Mining ◽  
Frequent Pattern ◽  
Frequent Patterns ◽  
Time Cost ◽  
Detection Approach ◽  
Maximal Frequent Pattern ◽  
Weighted Data

Since outliers are the major factors that affect accuracy in data science, many outlier detection approaches have been proposed for effectively identifying the implicit outliers from static datasets, thereby improving the reliability of the data. In recent years, data streams have been the main form of data, and the data elements in a data stream are not always of equal importance. However, the existing outlier detection approaches do not consider the weight conditions; hence, these methods are not suitable for processing weighted data streams. In addition, the traditional pattern-based outlier detection approaches incur a high time cost in the outlier detection phase. Aiming at overcoming these problems, this paper proposes a two-phase pattern-based outlier detection approach, namely, WMFP-Outlier, for effectively detecting the implicit outliers from a weighted data stream, in which the maximal frequent patterns are used instead of the frequent patterns to accelerate the process of outlier detection. In the process of maximal frequent-pattern mining, the anti-monotonicity property and MFP-array structure are used to accelerate the mining operation. In the process of outlier detection, three deviation indices are designed for measuring the degree of abnormality of each transaction, and the transactions with the highest degrees of abnormality are judged as outliers. Last, several experimental studies are conducted on a synthetic dataset to evaluate the performance of the proposed WMFP-Outlier approach. The results demonstrate that the accuracy of the WMFP-Outlier approach is higher compared to the existing pattern-based outlier detection approaches, and the time cost of the outlier detection phase of WMFP-Outlier is lower than those of the other four compared pattern-based outlier detection approaches.

Mining Closed Weighed Frequent Patterns from a Sliding Window over Data Stream

2013 ◽  
Vol 756-759 ◽  
pp. 2606-2609
Author(s):  
Cui Cui Ge ◽  
Xiu Fen Fu
Keyword(s):  
Data Stream ◽  
Pattern Mining ◽  
Frequent Pattern Mining ◽  
Sliding Window ◽  
Frequent Itemset ◽  
Experimental Results ◽  
Frequent Pattern ◽  
Frequent Patterns ◽  
Itemset Mining ◽  
Basic Window

Weighted frequent pattern mining address to discover more important frequent pattern by considering different weights of every item, closed frequent pattern mining can significantly reduce the number of frequent itemset mining and keep sufficient result information. In this paper,we proposed an algorithm DS_CRWF to mine closed weighted frequent pattern over data stream,which is based on sliding window and take basic window as unit of updating,all the closed weighted frequent patterns can be mined through once scan.The experimental results show the feasibility of the algorithm.

A Novel Algorithm for Predicting Valuable Items in Data Streams

2015 ◽  
Vol 3 (2) ◽  
Author(s):  
S. Vijayarani Mohan
Keyword(s):  
Data Mining ◽  
Data Streams ◽  
Data Stream ◽  
Pattern Mining ◽  
Research Work ◽  
Frequent Pattern Mining ◽  
Frequent Pattern ◽  
Frequent Items ◽  
And Performance ◽  
Transactional Data

A data stream is a real time, continuous, structured sequence of data items. Mining data stream is the process of extracting knowledge from continuous arrival of rapid data records. Data can arrive fast and in continuous manner. It is very difficult to perform mining process. Normally, stream mining algorithms are designed to scan the database only once, and it is a complicated task to extract the knowledge from the database by a single scan. Data streams are a computational challenge to data mining problems because of the additional algorithmic constraints created by the large volume of data. Popular data mining techniques namely clustering, classification, and frequent pattern mining are applied to data streams for extracting the knowledge. This research work mainly concentrates on how to predict the valuable items which are found in a transactional data of a data stream. In the literature, most of the researchers have discussed about how the frequent items are mined from the data streams. This research work helps to predict the valuable items in a transactional data. Frequent item mining is defined as finding the items which occur frequently, i.e. the occurrence of items above the given threshold is considered as frequent items. Valuable item mining is nothing but finding the costliest or most valuable items of a database. Predicting this information helps businesses to know about the sales details about the valuable items which guide to make crucial decisions, such as catalogue drawing, cross promotion, end user shopping, and performance scrutiny. In this research work, a new algorithm namely VIM (Valuable Item Mining) is proposed for finding the valuable items in data streams. The performance of this algorithm is analysed by using the factors, number of valuable items discovered, and execution time.

Sign in / Sign up

Export Citation Format

Share Document