flowchart and algorithm of electricity bill genration in c language

flowchart and algorithm of electricity billing system

ABSTRACT
The relationships among the classes play an important role for the selection of the object-oriented database. In this context, association rules show the relationship between the two classes. As the object-oriented design for the development of the software grows, the relationships among the classes also grow towards the complex software design but one can develop optimize object-oriented design through right association rules between classes. The present paper deals to design the right association rules for the object-oriented databases taken from the classes. A real case study of Electricity Bill Deposit System (EBDS) is considered in Indian scenario and Apriori algorithm is used for finding the use of frequent data sets through the right association rules. The association rules are designed through well-known Unified Modeling Language (UML). The present work is an implementation of Apriori algorithm towards the database of EBDS.

1.RELATED WORK
Association rules are used to find out the hidden relationship from the stored database. Association rule is a key concept of data mining technique for finding the hidden relationship. Association rules works like if and then statement. Association rules are dividing in two parts, one is antecedent other is consequent. An antecedent finds the items from the stored large data item while consequent is an item found in dataset and then combined with antecedent. Association rules are created by analyzing data for frequent item set. Support and Confidence are most important relationships in association rules. Support is an indication, how it appears in the database while Confidence shows how many times it is appearing in database. Let us first describe some of the important references related to the concepts of data mining. Agrawal et al. [1] have described the irregularities from the large scale transaction data generated at point-of-sale in supermarkets by the use of association rules. Sergey et al. [2] have considered the problem of analyzing market-basket data and they presented some contributions for finding large item sets based on sampling. Cai et al. [3] have proposed a technique, where the items are considered as per weight. The special promotions are given to the item which maximise the revenue of different items. Authors also described weighted association rules with the weight which presented the algorithm to explore the problem of down words closure property which support measure in the un-weighted case and the case is no longer exit. Dunkel and Sparkar [4] have described it as the main aspects namely representation organization and the next one is access of the particular data. When the loss of input and output are considered the access speed of data are significantly affect the performance. Authors also used one thing, comparison between column and row wise approaches and the data accessed through Apriori association algorithm and it is found that counting using Apriori algorithm with data accessed in column wise manner is better by reducing the degree to which the data as well as contents needs to be rapidly brought into the memory. Walter [5] has described the idea of tupple or element recording which helps to improve the low level efficiency of the algorithm. Burdick et al. [6] have described a depth first traversal of item set lattice and developed as a search as used as strategy algorithm which is the effective pruning mechanism for mining performance. A combination of vertical bitmap representation of data and efficient bitmap compression schema are described and dependent upon for implementation of support counting. Chai [7] has presented the advance Apriori algorithm which enhances the perfection of generating association rules in order to adopt new method to depth the association problem of redundant generation of rules itemsets. Taboada [8] has described the concept of data mining for extracting useful hidden knowledge from large volumes of data for their use in decision support system.Weng et al. [9] have described the relationships hidden in large data sets by association analysis for frequent item sets or association analysis which highlight the relationship to reduce support or confidence of the rules. The other important related reference on data mining is [10]. Umarani [11] has introduced identification of association rule in large databases in order to form the first step in mining of association rules in the process of frequent itemset mining. Verma [12] has described the database community mostly on the association rule of data mining. The rule seeks to correlate among attributes in a database and is important in generating frequent item sets from large database. Regular retrieval and storage of huge data have leaded the Industries to develop association rules from the database. He et al. [13] have described aggregate values of varying attributes computed by various systems by processing data in large databases and/or warehouses. In the present work, association rules have been designed to find out hidden relationship from a large amount of data stored in database. Object-oriented database has been used for the EBDS in Indian context. Association rules are represented through Unified Modeling Language class diagram. Apriori algorithm is used for finding the frequent used database from the large data sets and represented through diagram. The present work is useful when the size of database is growing day-by-day in many business organizations.

2. UNIFIED MODELING LANGUAGE FOR ASSOCIATION RULES

Unified Modeling Language (UML) represents the graphical view of the modeling language and the important feature of UML is to design the software models. It is very useful for analysis of any research problem and UML class diagram depicts the class name, attributes and operations. UML class diagram divides the data into three rows. First row shows the class name, second row shows the attribute name while third column shows the operations. The present work is a case study of EBDS and figure 1 shows the UML class representation of EBDS and it consists of relationship between Meter and Plan. The association is that the Meter distributed according to Plan. A customer depicts take a different type of plans depending on the condition. The UML class diagram as shown in figure 2 depicts the relationship between Meter, Plan and Meter_Reader. A customer depicts the electricity bill depending on the total reading. The UML class diagram as shown in figure 3 represents the relation between Meter, Plan, Meter_Reader and Bill. This diagram shows that customer may take any type of plan to deposit the electricity bill. However, the amount will inadvertently depend on the total reading.