Bio-Informatization and Application of Distributed Data Mining to Facilities Management
Ezendu I. Ariwa
Department of Accounting, Banking & Financial Systems
London Metropolitan University
United Kingdom
Mohamed M. Medhat
Sadat Academy for Management Sciences
Egypt
Introduction/Abstract
Distributed data mining has great functionalities that can offer to nowadays applications. That is because the nature of most of these applications is data distribution. One of the potential applications for distributed data mining is the use of OIKI DDM model in Facilities Management (FM). In this paper we investigate the potential advantages of this approach.
Keywords: Facilities Management – Knowledge Discovery – OIKI DDM – Decision support system.
1- Introduction
There is no agreed definition on the term “Facilities Management” in the literature. However we could define it as follows: Facilities Management (FM) is an integrated approach to operating, maintaining, improving and adapting the infrastructure of an organization in order to build an environment that strongly supports the primary objectives of the organization. The FM uses information in order to accomplish its task. This information inherently distributed among a number of heterogeneous databases in different loosely coupled sites connected by a computer network [5].
Distributed data mining refers to
the mining of distributed data sets. The data sets
are stored in local databases, hosted by local com
puters, which are connected through a computer net
work. Data mining takes place at a local level and
at a global level where local data mining results are
combined to gain global findings [11].
In some applications, data are inherently dis
tributed, but it is necessary to gain global in
sights from the distributed data sets. For exam
ple, each site of a multinational company man
ages its own operational data locally, but the
data must be analyzed for global patterns to al
low companywide activities such as planning,
marketing, and sales. One of the direct applications of distributed data mining is the use of it in FM in order to improve the decision making process.
2- OIKI DDM Model
Senousy and Medhat have proposed Optimized Incremental Knowledge Integration (OIKI) DDM model, it is a mobile agent based DDM model that overcome the drawbacks of the traditional mobile agent based DDM models. Instead of transferring the results from each data server to the client, the client controls migration of the results among data servers to be integrated locally and finally, the final results are transferred to the client [11].
The typical OIKI DDM process: the client multicasts MADMs and MAKIs (Mobile Agents-Knowledge Integrators) to the required data servers. The data mining process is performed locally on each data server. The size of results of the first two accomplished DM processes are compared. The smaller one is migrated to the larger one. The knowledge integrator agent integrates the results of these two data servers. This process is repeated until all integrated results are resident in a specific data server and finally, the final results are sent back to the client. Consequently, the OIKI DDM process passes through three main stages: 1) Preparation Stage: The client multicasts MADMs and MAKIs to data servers. 2) Data Mining Stage: Data mining process is performed locally on each data server. 3) Knowledge Integration Stage: An incremental knowledge integration technique is performed on the data servers where the smaller results are migrated to the larger one to optimize the cost of results migration among data servers.
The client generates DM request, then it determines the required data servers needed in the DM request and sends MADMs and MAKIs to the specified data severs. Each data server sends size of the generated results and the timestamp upon accomplishing the data mining process back to the client. The client arranges the data servers’ results information in a queue according to the timestamp of each. The client makes a comparison between two consecutive items in the queue according to the result size and sends a control command to the smaller result to migrate to the larger one. The knowledge integration process takes place at the data server that has the larger result size. The loop continues till all the results are integrated and are resident in one data server. Finally, the data server that contains the final results sends it back to the client.
3- Application of OIKI DDM model to FM
A typical application of OIKI DDM model to FM that MADMs and MAKIs visit each FM used database: fleet DB, catering DB, print DB, etc in order to deduce the hidden knowledge from these distributed databases in the same operation steps illustrated above. Figure (1) illustrates this application.
…
From the literature, we can deduce some potential advantages of our approach. These potential advantages could be summarized as follows:
1- Since distributed data mining could result in a global view of the distributed data, the effect of each information unit of an organization FM related data could be discovered. [1, 6, 8].
2- This approach is cheaper than the data warehousing approach since the data warehouse creation is not necessary to accomplish the OIKI DDM process [11].
3- The use of mobile agents in this context could adapt the system to use e-FM using Intranet within the organization [4, 13].
4- The adoption of mobile agents could adapt the system to use extranet in order to integrate among FM of the organization and its partners [2, 4, 7,10, 12].
5- The mobile agents can roam the Internet to deduce some of the hidden knowledge in order to make the FM decision according to global market conditions [3, 9, 13].
4- Conclusions and Future Work
The use of OIKI DDM model in Facilities Management would increase the efficiency of the decision support system of that sector. The potential advantages of the model have been discussed supported from the literature. Our future work is to apply this model on some organizations in the UK and Egypt.
References
[1] R. Agrawal, T. Imielinski and A. Swami. "Mining Association Rules Between Sets of Items in Large Databases", in Proc. of the ACM Int. Conf. on Management of Data, Washington, USA, May 1993.
[2] D. Chess, C. Harrison et A. Kershenbaum. Mobile Agents: Are They a Good Idea ?. IBM Research Division, T.J. Watson Research Center, Yorktown Heights, New York, march 1995.
[3] Further Education Funding Council (1997) Effective Facilities Management: A Good Practice Guide. Her Majesty's Stationery Office.
[4] E. Garcia-Lopez, Distributed Management Facilities Architecture, TINA-C baseline document TB_EGL.002_2.1_1996.
[5] Graham Goulbourne and Frans Coenen and Paul H. Leng, "KD in FM: Knowledge Discovery in Facilities Management Databases", Database and Expert Systems Applications. pp. 806-815, 1998.
[6] R. Grossman, S. Bailey, S. Kasif, D. Mon, A. Ramu, and B. Malhi. Design of papyrus: A system for high performance, distributed data mining over clusters, meta-clusters and super-clusters. In Proceedings of Workshop on Distributed Data Mining, along with KDD98, Aug 1998.
[7] James E. White. Mobile Agents. In Jeffrey Bradshaw, editor, Software Agents. The MIT Press, 1996.
[8] H. Kargupta and P. Chan, editors. Advances in Distributed Data Mining. AAAI Press, 2000.
[9] Kendall, E. A., P.V. Murali Krishna, Chirag V. Pathak, C.B. Suresh, "Patterns of Intelligent and Mobile Agents," Agents '98, May, 1998.
[10] Ramu,A,T., (1998), "Incorporating Transportable Software Agents into a Wide Area High Performance Distributed Data Mining Systems", Masters Thesis, University of Illinois, Chicago, USA.
[11] M. Senousy, and M. Medhat. A proposed model for Distributed Data Mining using Mobile Agents. BIT 2001 “Constructing IS Futures,” Manchester. UK. 2001.
[12] Subramonian, R., & Parthasarathy, S. (1998). An architecture for distributed data mining. Fourth International Conference of Knowledge Discovery and Data Mining, New York, New York, Pages 44--59.
[13] Wu Z. D. and D. J. Hughes, "An Approach to Integrate Management Facilities for Campus Network Environments", Proceedings of the 1991 Singapore International Conference on Networks, pp. 131-136, September 1991.
1