Bhawna Mallick

Incremental mining of sequential patterns: Progress and challenges

Sequential pattern mining is a vital problem with broad applications. However, it is also challenging, as combinatorial high number of intermediate subsequences are generated that have to be critically examined. Most of the basic solutions are based on the assumption that the mining is performed on static database. But modern day databases are being continuously updated and are dynamic in nature. So, incremental mining of sequential patterns has become the norm. This article investigates the need for incremental mining of sequential patterns. An analytical study, focusing on the characteristics, has been made for more than twenty incremental mining algorithms. Further, we have discussed the issues associated with each of them. We infer that the better approach is incremental mining on the progressive database. The three more relevant algorithms, based on this approach, are also studied in depth along with the other work done in this area. This would give scope for future research direction.

CRM Customer Value based on Constrained Sequential Pattern Mining

role of data mining has become increasingly important for an organization that has large databases of information on customers. Customer Relationship Management (CRM) systems are implemented to identify the most profitable customers and manage the relationship of company with them. Intelligent data mining tools and techniques are used as backbone to the whole CRM initiative taken by the companies. Data mining tools search the data warehouse maintained by the companies and predict the hidden patterns and present them in form of a model. Strategic decisions about the customers can be taken based on the outcomes of these models. The data mining researchers have presented various mining algorithms to extract patterns in data for successful CRM approach. These approaches are however facing several problems like they are not business-focused and often results in enormous size of data after applying mining approaches. They have no relevant mechanism to provide guidance for focusing on specific category of customers for business profitability. In this article, the requirements of sequential pattern mining process for CRM environment is described, and then a novel constraint guided model for knowledge discovery process is proposed. We have suggested even how the selection of appropriate constraints can be made from the perspective of customer value analysis.

Progressive CFM-Miner: An Algorithm to Mine CFM – Sequential Patterns from a Progressive Database

Sequential pattern mining is a vital data mining task to discover the frequently occurring patterns in sequence databases. As databases develop, the problem of maintaining sequential patterns over an extensively long period of time turn into essential, since a large number of new records may be added to a database. To reflect the current state of the database where previous sequential patterns would become irrelevant and new sequential patterns might appear, there is a need for efficient algorithms to update, maintain and manage the information discovered. Several efficient algorithms for maintaining sequential patterns have been developed. Here, we have presented an efficient algorithm to handle the maintenance problem of CFM-sequential patterns (Compact, Frequent, Monetaryconstraints based sequential patterns). In order to efficiently capture the dynamic nature of data addition and deletion into the mining problem, initially, we construct the updated CFM-tree using the CFM patterns obtained from the static database. Then, the database gets updated from the distributed sources that have data which may be static, inserted, or deleted. Whenever the database is updated from the multiple sources, CFM tree is also updated by including the updated sequence. Then, the updated CFM-tree is used to mine the progressive CFM-patterns using the proposed tree pattern mining algorithm. Finally, the experimentation is carried out using the synthetic and real life distributed databases that are given to the progressive CFM-miner. The experimental results and analysis provides better results in terms of the generated number of sequential patterns, execution time and the memory usage over the existing IncSpan algorithm.