Journal of Global Research in Computer Sciences
MenuISSN: 2229-371X
ISSN: 2229-371X
Ashu Singla*1, Raman Maini*2
|
| Related article at Pubmed, Scholar Google |
Visit for more related articles at Journal of Global Research in Computer Sciences
Project assigning is a major decision for any software organization like IT companies to their developer employees. The uncertain domain of assessment has been in need of a reliable and consistent system to help simplify the decision making process. Logic rules provides a completely different, unorthodox way to approach a control problem. This method focuses on what the system should do rather than trying to understand how it works. One can concentrate on solving the problem rather than trying to model the system mathematically,if that is even possible. This almost invariably leads to quicker, cheaper solutions. This research work will help in analysis assign project to a developer according to different factors like reliability,his hardware interaction, programming level,configuration etc the decision factor helps in making decision above how to select accurate developer for the project. Using the algorithm has been defined, new rules defined,according to different rules added and take average decision based on data collected by different categories of the employees.
Keywords |
| Logic rules, Decision support system, Visual basic,Iiterative Dichotomiser3 |
INTRODUCTION |
| It is very difficult to evaluate factors for any organization to assign the projects. The main problem of the research work is that to evaluate different factors about a new employee (developer). So in very short time it is difficult to analyze to whom we will assign the project or whom not. As human brain is capable of analyzing few factors regarding new employee, because every case is new or different even for average experienced person. The question though is how to resolve the pressure from human perception in making a judgment. Data mining is making changes to the entire make up of our skills and comfort zones in information analysis [1]. Data mining is a new type of exploratory and predictive data analysis whose purpose is to establish systematic relations between variable when there are no prior expectations as per to nature of those relations. Data mining is a process of extracting previously unknown, valid, potentionally useful and hidden patterns from large datasets. As the amount of data stored in company databases is increasing rapidly, in order to get required benefits from large data sets and to find hidden relationships between variables using different data mining techniques developed and used. Clustering and decision tree are most widely used techniques for future predictions. Decision trees bare analytical tools used to discover rules and relationships by systematically breaking down and sub dividing the information contained in data set. Decision trees are composed of a hierarchy of “ifthen” statements and thet are most appropriate for categorical and interval data. |
| Since the data set of possible decision trees is too large to searched exhaustively, recursive algorithms were constructed during the last 10 years, for e.g., SLIQ (supervised learning in Quest) algorithm developed by IBM’s Quest project team in which pre-sorting is done in the tree-growth to later on avoid a costly sorting at each node. Parallel algorithms for data mining have also been suggested by a group of computer scientists. In this approach n training data are randomly distributed to P processors. Two new methods were developed for this purpose: the synchronous Tree construction approch and the and the partitioned tree construction approch [6]. Traditionally approch attempts to develop analysis system based on the logic using two structures: traditional reasoning of all inputs that map to one single output and stage wise reasoning of input parameters in accordance with their importance. These models are advanced but sometimes complex and can only be understood by specialists [8][9]. The expert rules were constructed using the reasoning in order to adequately analysed the inputs. This paper investigates way of logic implementation for a project assigns predictor. It’s a basically decision support system which helps in decision-making regarding a new project assignment to the developer. The approch considers all risk influencing input parameters in single stage of the decision making process. Member functions have been plot according to the different input variables and their rules are defined. From different rules added, decision has been taken according to decision as accepted or rejected. For accurate results studies are presented. |
| The rest of this paper is organized as follows. Section 2 deals with the overview of the research work, section 3 and 4 describes about the small introduction of decision trees and ID3 algorithm. Section 5 is having the results and the last section concludes the finding of the proposed research work and future or scope of the research work. |
OVERVIEW OF THIS WORK |
| In company’s database there are information about each developer like their name, their programming level, their compatibility with the configuration, how much a person can be reliable to a company, and he or she can interact with the hardware or not, the language in which they can perform well and many more. The proposed model makes prediction about when a new developer is recruited then which project is assigned to him or her. It is preliminary that the assigned new project will be depend on the qualities as attributes which he has but the previous data will help to take a decision in which area he or she should be assigned. Following the methodology shown the stages involved in the process of this work. Mainly the input data is taken and take it in the required format as the software needs then applied the ID3 algorithm. Further ID3 takes many steps and then the output will be the logic rules, which will be the result. |
 |
DECISION TREES |
| A decision tree is a tree in which each branch represents a choice variant, and each leaf node is a decision. This kind of trees is often used with the object of information gain in a taking decision process. [2] Decision trees are used to solve the problems with predictions; these basically help in solving the decision oriented problems, where generally fuzziness is there. The basic objective of this work is to study the problem solving capability of decision tree. Decision tree are commonly used for gaining information for the purpose of decision making. Decision tree starts with a root node on which it is for users to take actions. From this node, users split each node recursively according to decision tree learning algorithm. The final result is a decision tree in which each branch represents a possible scenario of decision and its outcome.Learning about decision trees is a method of discreet value approximate, in which the learning function is a decision tree. This method is one of the most used techniques of inductive inference.[5][7] Decision tree induction is closely related to rule induction. Each path from the root of a decision tree to one of its leaves can be transformed into a rule simply by conjoining the tests along the path to form the antecedent part, and taking the leaf’s class prediction as the class value. |
ID3 ALGORITHM BASIC |
| ID3 is a simple decision tree-learning algorithm developed by Ross Quinlan (1983). The basic idea of ID3 algorithm is to construct the decision tree by employing a top-down, greedy search through the given sets to test each attribute at every tree node [3][4]. A measure called information gain, which will be used to decide which attribute to test at each node. Information gain is itself calculated using a measure called entropy, which we first define for the case of a binary decision problem and then define for the general case. Given a binary categorization, C, and a set of examples, S, for which the proportion of examples categorized as positive by C is P(positive)+ and the proportion of examples categorized as negative by C is P(negative), then the entropy of S is: |
| Entropy(S)= -P(positive)log2P(positive) – P(negative)log2P(negative) (i) |
| P(positive): proportion of positive examples in S |
| P(negative): proportion of negative examples in S |
| Note that the more uniform is the probability distribution, the greater is its information. You may notice that entropy is a measure of the impurity in a collection of training sets. Information Gain: We now return to the problem of trying to determine the best attribute to choose for a particular node in a tree[10]. The following measure calculates a numerical value for a given attribute, A, with respect to a set of examples, S. Note that the values of attribute A will range over a set of possibilities which we call Values (A), and that, for a particular value from that set, v, we write Sv for the set of examples which have value v for attribute A. The information gain of attribute A, relative to a collection of examples, S, is calculated as: |
| Gain (S, A) = Entropy (S) - Sum for v from 1 to n of (|Sv|/|S|) * Entropy (Sv) |
| The algorithm terminates either when all the attributes have been exhausted, or the decision tree perfectly classifies the problem. It uses information gain to measure the attribute to put in each node, and performs a greedy search using this measure of worth. The algorithm goes as follows: |
| Given a set of examples, S, categorized in categories ci, then: |
| 1. Choose the root node to be the attribute, A, which scores the highest for information gain relative to S. |
| 2. For each value v that A can possibly take, draw a branch from the node. |
| 3. For each branch from A corresponding to value v, calculate Sv. Then: |
| If Sv is empty, choose the category cdefault , which contains the most examples from S, and put this as the leaf node category which ends that branch. |
| If Sv contains only examples from a category c, then put c as the leaf node category, which ends that branch. |
| Otherwise, remove A from the set of attributes which can be put into nodes. Then put a new node in the decision tree, where the new attribute being tested in the node is the one which scores highest for information gain relative to Sv. This new node starts the cycle again, with S replaced by Svin the calculations and the tree gets built iteratively like this. |
DESIGN OF THE PREDICTOR TOOL AND RESULTS |
| In the process of designing a predictor the most important task is to identify those factors that contribute primarily to a software company’s decision concerning providing projects to a right person for developing. In order to identify the process and influencing the factors those contributes to a developer’s assessments; the analysis work of an experienced manager in a software company was observed. After discussion we identified the main factors are programming level, configuration, reliability, hardware interaction, computer language etc are important factors on which theis decision can be based. |
 |
| Here the technical terms are taken by the industry for the programming level such as Terran for the technical leader with exp of 10-15 years,Zerg as on particular platform or any particular tool like oracle, mysql and Protoss use algo’s and mathematical tools. The impact levels of various parameters can be set or change according to the system of the different organizations [11][12]. The GUI based tool is developed according to different rules added, which is developed using Visual Basic language as shown in fig2 and corresponding decision by this tool in fig 3 |
 |
 |
CONCLUSION |
| A logic-based predictor is developed to assist Software Company in decision-making. Decision tree learning algorithm has been successfully used in expert systems in capturing knowledge. The main task performed in these systems is using inductive methods to the given values of attributes of an unknown object to determine appropriate classification according to decision tree rules. . A simple GUI based applicatin, which is developed using visual basic, is simple and easy to use. The main focus of this depend upon the impact factors of attributes, it is very clear to analysis decision whether a project is assign or not for this application. From this research paper, another students can take the idea about the application of data mining in new research areas. |
ACKNOWLEDGMENT |
| I would like to thank my Research Guide Raman Maini, Associate Professor, University College of Engineering, Punjabi University Patiala for his valuable assistance, help and guidance during the research process. |
References |
|