Bai 2008 Myanmar Paper

IMPLEMENTING A DECISION SUPPORT SYSTEM (DSS) FOR DISCRIMINATING HERBAL AND MEDICINAL PLANTS IN MYANMAR USING CLASSIFICATION TECHNIQUE Hay Mar Hlaing Pakokku Computer University, Myanmar [email protected] Thida Oo Pakokku Computer University, Myanmar [email protected] Aung Kyaw Oo Information Technology Policy Program (ITPP), Technology Management, Economics and Policy Program (TEMEP), Seoul National University, San 56-1, Silim Dong, Gwan-Ak Gu, Seoul 151-742, Republic of Korea [email protected]

ABSTRACT Decision Support System (DSS) supports users help make decisions. DSS is built in a variety of ways to classify data. In this paper, DSS built for the classification among herbal and medicinal plants is discussed. Firstly, plants collected in the Botany Department of a University will be listed. Secondly, the herbal and medicinal plants among those will be picked and their data will be stored. The proposed system collects data of the commonly used herbal plants, classify them and provide the information of traditional and medicinal importance. It is implemented by classification technique in data mining and Java programming. Keyword: DSS, classification, herbal plants, traditional medicine 1. INTRODUCTION Nowadays, Information and Communication Technology (ICT) is piercing every-where in our everyday life and plays an important role in the fields related to computer science and engineering. Information Science can provide some solutions and profound supports for professionals, researchers and enthusiasts and end-users in academia, business and management.

Traditional medical plants are systematically collected in the Department of Botany of this University. These plants are in some ways or another essential for the health of people. In this paper, (120) kinds of traditional herbal plants among others are listed and also categorized in the

database by ways to use, its usage, its growing place, its name, its botanical name, its feature, its effect and family of each plant. Decisions about traditional herbal plants are to be made by the proposed DSS. The major purpose of the paper is to help people access the information about herbal plants they want to know, in other words, or to get the information of the herbal plants at hand and accurately. The paper is composed of seven sections. In Section (2), main objectives behind the paper are presented. Some theoretical things are shown in section (3). The next two sections, i.e. section (4) and (5), explain the related and specific theories about the present paper. Section (6) elaborates the system architecture of the proposed DSS. Conclusion is to be seen in section (7).

2. OBJECTIVES The objectives behind the paper are to:- know the herbal plants of traditional medicine for any users, - support the users by providing several guidance and suggestions upon traditional plants for curing various diseases,

medicinal

- collect the information about the commonly used herbal plants, and - implement a system to classify the herbal plants based on traditional medicine 3. THEORETICAL BACKGROUND 3.1 Data mining

Data mining is the extraction of implicit information from a large data set. The basic idea is that if the decision support data is not well understood, we can use the computing power of computers to help us discover patterns in the data. Also, data mining is the process of discovering interesting knowledge from large amounts of data stored either in database, data warehouses or other data repositories. The primary data-mining tasks are:1. Classification 2. Regressing

3. Clustering 4. Summarization 5. Dependency Modeling 6. Change and Deviation Detection

3.2 Classification

Classification is a form of data analysis that extracts a model from data to classify future data. It has been studied in parallel in statistics and machine learning, and is currently a major technique in data mining with a broad application spectrum. Since many application problems can be formulated as a classification problem and the volume of the available data has become overwhelming, developing scalable, efficient, domain-specific, and privacy-preserving classification algorithms is essential. 4. INFORMATION SYSTEM An information system (IS) collects, processes, stores, analyze and disseminates information for a specific purpose. Like any other systems, an information system includes input (data, instructions) and outputs (reports, calculations). It processes the inputs and produces outputs that are sent to the user or to other systems. 4.1 Types of Information System

Information system can be classified as follows:- Transaction processing system (TPS) - Management information system (MIS) - Office automation system (OAS) - Decision support system (DSS) - Executive information system (EIS)

- Group support system (GSS) - Intelligent support system (ISS) 4.2 Decision-making

Information is used to make decisions. Decision making is not a single activity that takes place all at one. The process consists of several different activities that take place at different times. The decision maker has to perceive and understand problems. Once perceived, solutions must be designed; once solutions are designed, choices have to be made about a particular solution; finally, the solution has to be carried out and implemented. Four different stages in decision making are intelligence, design, choice and implementation.

4.3 Decision Support System

A decision-support system is an integrated set of computer tool that allows a decision maker to interact directly with computers to create information useful in making semi-structured and unstructured decisions. The software components for decision-support systems are a language system which enables the user to interact with the decision-support system, a problem-processing system which is made up of several components that perform various processing tasks and a knowledge system which provides data and artificial-intelligence capabilities to the decision-support system. 5. IMPLEMENTATION OF THE SYSTEM USING CLASSIFICATION METHOD In implementing this system, a database of herbal plants on the traditional medicine is used. The database is used to send out learning. The database describes attributes of the herbal plants, such as their usage, botanical name, growing place, feature, effect and family of each plant. In this system, decision tree algorithm is used together with an entropy method, i.e. entropy-based measure known as information gain. 5.1 Attribute Selection Measure (ASM)

The information gain measure is used to select the test attribute at each node in the decision tree. Such a measure is referred to as an (ASM) or a measure of the goodness of split. The attribute with the highest information gain is chosen as the test attribute for the current node. Let S be a set consisting of S data samples. The expected information needed to classify a given sample is given by:

I (S1, S2, ---, Sm) = -ΣPiLog2 (Pi)…………. (1)

Where, Pi =Probability that an arbitrary sample belong to class Ci and is estimated by Si / S. Log2 = Log function to the base 2 The entropy or expected information based on the partitioning into subsets by A is given by equation (2):

E (A) =

ν S1j ,......., Smj ∑ S j=1

(S1j,......., Smj)……. (2)

The term acts as the weight of the jth subset and is the number of samples in the subset divided by the total number of samples in S.

The encoding information that would be gained by branching on A is given by: Gain (A) =I (S1, S2, -----, Sm) - E (A)………… (3) In other words, Gain (A) is the expected reduction in entropy caused by knowing the value of attribute A. The algorithm computes the information gain of each attribute. 6. SYSTEM ARCHITECTURE This system includes user, ( 120 ) kinds of herbal plants and their all information such as usage, growing place (habitat), botanical name, feature, effect, the way to use and family (superclass and subclasses) of each plant. The system will check and search the plant name in the

database. If the search matches with the data in the database, the user will be shown the plant name that is extracted from the database. If a user wants to search plant name that does not exist in the database, the system returns to the input stage. If the search data exists in the database, the plants are classified by usage or symptom or skeleton. Finally, the information is displayed to the user who wants to know.

User

New

Input

Search No Search Plant

Name/Symptom/S keleton

Yes Classify Process

Database

Usage/Symptom / Skeleton

Display

Figure (6.1): Data Flow Diagram

7. CONCLUSION

In this paper, an information system that provides information for decision making, i.e., Decision Support System, about the collection of commonly used herbal plants, providing the information of traditional medicine and classify those plants are discussed. As it showed in the previous sections, Data mining is the task of discovering interesting patterns from large amounts of data set. The process discussed in this paper has many advantages. This system can support to help decide the user the nature of herbal plants and can discriminate the categories of the traditional medicine concerned as well. Last but not least, the proposed system in this paper can be used as a reliable reference for herbal plants and traditional medicine in Myanmar.

REFERENCE [1]

D. B. Leake. (Ed.), Cased-Based Reading: Experience, Lessons and Future Directions. Menlo Park: AAA I Press, 1996.

[2]

U.M. Fayyad, G. Piatestky Shapiro, P.Smyth and R. Uthurusamy, (Eds), Advance in Knowledge Discovery and Data Mining. Cambridge, MA: MIT Press, 1996.

[3]

D. Gibson, J. M. Kleinberg, and P. Raghavan., Clustering Categorical Data: An Approach based on Dynamical Systems, Proceedings of International Conference on Very Large Data Bases( VLDB’98), New York, Aug 1998.

[4]

G. Piatetsky Shapiro and W. J. Frawley, (Eds.), Knowledge Discovery in Databases. Cambridge, MA: AAAI/ MIT Press, 1991.

[5]

T. Fukuda, Y. Morimoto, S. Morishita, and T. Tokuyama., Data mining using two-dimensional optimized association rules: Scheme, algorithms and visualization, Proceedings of ACMSIGMOD Int’l. Conference on Management of Data (SIGMOD 96), Montreal, Canada, June 1996.

[6]

Vladimir., Electronic Commerce: Structure and Issues, International Journal of Electronic Commerce, 1, No.1, Fall 1996.

[7] [8]

Herbert A., The New Science of Management Decision, New York: Harper & Row, 1960. Steven L., Decision Support System: Current Practice and Continuing Challenges, Reading, MA: Addison -Wesley, 1980.