Seminar Nasional Teknologi Informasi 2016

Similar documents
CHAPTER 8 DECISION SUPPORT V2 ADVANCED DATABASE SYSTEMS. Assist. Prof. Dr. Volkan TUNALI

This tutorial will help computer science graduates to understand the basic-to-advanced concepts related to data warehousing.

Power Distribution Analysis For Electrical Usage In Province Area Using Olap (Online Analytical Processing)

A Novel Approach of Data Warehouse OLTP and OLAP Technology for Supporting Management prospective

Data Mining Concepts & Techniques

Rocky Mountain Technology Ventures

RETRACTED ARTICLE. Web-Based Data Mining in System Design and Implementation. Open Access. Jianhu Gong 1* and Jianzhi Gong 2

REPORTING AND QUERY TOOLS AND APPLICATIONS

DATA MINING AND WAREHOUSING

CSPP 53017: Data Warehousing Winter 2013! Lecture 7! Svetlozar Nestorov! Class News!

Adnan YAZICI Computer Engineering Department

International Journal of Scientific & Engineering Research, Volume 7, Issue 11, November ISSN

COMM 391 Winter 2014 Term 1. Tutorial 1: Microsoft Excel - Creating Pivot Table

collection of data that is used primarily in organizational decision making.

Data Warehousing and OLAP Technology for Primary Industry

OLAP Introduction and Overview

DATA WAREHOUING UNIT I

Information Integration

Question Bank. 4) It is the source of information later delivered to data marts.

Implementing Data Models and Reports with SQL Server 2014

The application of OLAP and Data mining technology in the analysis of. book lending

CHAPTER 8: ONLINE ANALYTICAL PROCESSING(OLAP)

CT75 DATA WAREHOUSING AND DATA MINING DEC 2015

Decision Support, Data Warehousing, and OLAP

Chapter 3. Databases and Data Warehouses: Building Business Intelligence

Table of Contents. Knowledge Management Data Warehouses and Data Mining. Introduction and Motivation

Knowledge Management Data Warehouses and Data Mining

Step-by-step data transformation

The strategic advantage of OLAP and multidimensional analysis

ETL and OLAP Systems

CS377: Database Systems Data Warehouse and Data Mining. Li Xiong Department of Mathematics and Computer Science Emory University

Business Analytics Enhancements

Data Mining & Data Warehouse

MOLAP Data Warehouse of a Software Products Servicing Call Center

IT DATA WAREHOUSING AND DATA MINING UNIT-2 BUSINESS ANALYSIS

Data Warehousing and Decision Support

On the Integration of Autonomous Data Marts

Knowledge Modelling and Management. Part B (9)

WKU-MIS-B10 Data Management: Warehousing, Analyzing, Mining, and Visualization. Management Information Systems

Chapter 18: Data Analysis and Mining

Data Warehousing & Mining. Data integration. OLTP versus OLAP. CPS 116 Introduction to Database Systems

Q1) Describe business intelligence system development phases? (6 marks)

Aggregating Knowledge in a Data Warehouse and Multidimensional Analysis

Data Warehousing and Decision Support. Introduction. Three Complementary Trends. [R&G] Chapter 23, Part A

Data Warehousing. Ritham Vashisht, Sukhdeep Kaur and Shobti Saini

Data Warehousing and OLAP

Evolution of Database Systems

Viságe.BIT. An OLAP/Data Warehouse solution for multi-valued databases

Data Warehousing and OLAP Technologies for Decision-Making Process

1. Inroduction to Data Mininig

CT75 (ALCCS) DATA WAREHOUSING AND DATA MINING JUN

Search Of Favorite Books As A Visitor Recommendation of The Fmipa Library Using CT-Pro Algorithm

Data Analysis and Data Science

GUJARAT TECHNOLOGICAL UNIVERSITY MASTER OF COMPUTER APPLICATIONS (MCA) Semester: IV

Optimization Online Analytical Processing (OLAP) Data Sales Door Case Study CV Adilia Lestari

CHAPTER 3 Implementation of Data warehouse in Data Mining

Basics of Dimensional Modeling

On-Line Analytical Processing (OLAP) Traditional OLTP

An Overview of Data Warehousing and OLAP Technology

Database design View Access patterns Need for separate data warehouse:- A multidimensional data model:-

BUSINESS INTELLIGENCE FOR EVALUATION E-VOUCHER AIRLINE REPORT

Data Warehousing and Decision Support

Data Warehouse and Data Mining

Chapter 1, Introduction

Slice Intelligence!

Implementing and Maintaining Microsoft SQL Server 2008 Analysis Services

Guide Users along Information Pathways and Surf through the Data

Information Management course

MIS2502: Data Analytics Dimensional Data Modeling. Jing Gong

Data Modeling and Databases Ch 7: Schemas. Gustavo Alonso, Ce Zhang Systems Group Department of Computer Science ETH Zürich

Data Warehouses. Yanlei Diao. Slides Courtesy of R. Ramakrishnan and J. Gehrke

DATA MINING TRANSACTION

COURSE 20466D: IMPLEMENTING DATA MODELS AND REPORTS WITH MICROSOFT SQL SERVER

Data warehouses Decision support The multidimensional model OLAP queries

Information Management course

Multi-dimensional database design and implementation of dam safety monitoring system

A Benchmarking Criteria for the Evaluation of OLAP Tools

SQL Server Analysis Services

Constructing Object Oriented Class for extracting and using data from data cube

SCHEME OF COURSE WORK. Data Warehousing and Data mining

Overview. Introduction to Data Warehousing and Business Intelligence. BI Is Important. What is Business Intelligence (BI)?

Horizontal Aggregations in SQL to Prepare Data Sets Using PIVOT Operator

Decision Support. Chapter 25. CS 286, UC Berkeley, Spring 2007, R. Ramakrishnan 1

Deccansoft Software Services Microsoft Silver Learning Partner. SSAS Syllabus

Fig 1.2: Relationship between DW, ODS and OLTP Systems

Chapter 13 Business Intelligence and Data Warehouses The Need for Data Analysis Business Intelligence. Objectives

A Proposal of Integrating Data Mining and On-Line Analytical Processing in Data Warehouse

1 Dulcian, Inc., 2001 All rights reserved. Oracle9i Data Warehouse Review. Agenda

Visit our Web site at or call to learn about training classes that are added throughout the year.

Xcelerated Business Insights (xbi): Going beyond business intelligence to drive information value

International Journal of Computer Science & Information Technology Research Excellence Vol. 2, Issue 1, Jan-Feb 2011

INSTITUTE OF AERONAUTICAL ENGINEERING (Autonomous) Dundigal, Hyderabad

Data Warehousing and Data Mining. Announcements (December 1) Data integration. CPS 116 Introduction to Database Systems

20466C - Version: 1. Implementing Data Models and Reports with Microsoft SQL Server

CSE 544 Principles of Database Management Systems. Alvin Cheung Fall 2015 Lecture 8 - Data Warehousing and Column Stores

IMPLEMENTING STATISTICAL DOMAIN DATABASES IN POLAND. OPPORTUNITIES AND THREATS. Central Statistical Office in Poland

A Data Warehouse Implementation Using the Star Schema. For an outpatient hospital information system

The University of Iowa Intelligent Systems Laboratory The University of Iowa Intelligent Systems Laboratory

DATA WAREHOUSE EGCO321 DATABASE SYSTEMS KANAT POOLSAWASD DEPARTMENT OF COMPUTER ENGINEERING MAHIDOL UNIVERSITY

Data Warehouse and Data Mining

Transcription:

SMART ONLINE ANALYTICAL PROCESSING USING LIGHT CUBICAL DATA FOR MIDSIZE CORPORATION TO BUILD SHORT TERM BUSINESS STRATEGY EXTENDS WITH M5P PREDICTION ALGORITHM Feri Sulianta 1) Eka Angga Laksana 2) 1) 2) Teknik Informatika Universitas Widyatama Jl. CikutraNo. 204 A, Bandung, Indonesia email : 1) feri.sulianta@widyatama.ac.id, 2) eka. angga@widyatama.ac.id ABSTRACT The data in Midsize corporation will be empowered further for distinctive purposes and not just as transaction archives needs. Different kind of valuable information may be disclosed from the data transaction's sales generated from transaction processing system. In this case to clarify that involve thorough information that also involves diverse viewpoints are treated with a technique SMART Online Analytical Processing (OLAP). This technique is able to accommodate the completeness of the cube data, which is lightweight and compact to be analyzed in executive level. Referring to the restaurant transactional data, every consumer has a variety of options and different backgrounds which influence them for buying foods. OLAP techniques capable of presenting food transaction data in a multidimensional form. Furthermore, OLAP will execute the data that is slicing and dicing that summarize and collect large amounts of data, perform filtering, sorting, and rank that will enrich the valuable findings from the data cubes. The OLAP method will be combined with prediction using M5P Algorithm to extend the reliability for the business action that would be performed and give more information to the executive level to take advantage from OLAP system. Keywords- Online Analytical Processing, OLAP, Transaction Data, Midsize Corporation, Cubical Data, Slicing, Dicing, Business Strategy, Prediction, M5P Algorithm. I. INTRODUCTION A food industry have sales store and to run the business, company used transaction processing system. In certain period of time the data is accumulated in the database. The transactional data will be analyzed to reveal valuable information to build a business strategy, provide basic services and policy-making in the provision of products In this case, the company placed the data archives on data warehouse, which is useful to retrieve previous transaction [1][2][3][14]. Some terms of the information that can be explored, for example, the amount of raw material available because the product is not sold, look for products that consumers demand, settings and conditions affecting consumers into buying the product [4][10]. In this case the Online Analytical Processing techniques will be applied to systematization databases, data analysis and reveal valuable findings.vpractically, Online Analytical Processing techniques can bring the needs of the food product sales data retrieval.this method is expected to reveal a lot of information that was previously not addressed if only relies on the sales system with the features of a transaction processing system only. II. METHODOLOGY Online Analytical Processing, which is called OLAP, is a method of analyzing multidimensional data (or data set) interactively without having to use complicated programming techniques. Users simply access the graphics features available in the OLAP. Analysis of the data becomes easy; users can easily change and reposition the data with interactive graphics features. OLAP is mostly used in the manufacture of business reports such as financial statements, sales, marketing, predict the data, budget analysis, business process management, and extends the research and industrial purposes. OLAP enables analysts to analyze a variety of data systems with a variety of attributes, such as sales data, based on time of purchase, type of product, taste, and then compare it with the price of the products available data. Findings such as peak time of purchase and the most wanted products, or even the kind of food that does not sell, accumulated or unused product. Such facts can be considered as basis for making business strategy. For example, setting up more personnel in certain times were found to be the peak of the purchase, or even for business strategy by creating new products that consists of consumer preferred taste[16][17][19]. 175

Seminar Nasional Teknologi Informasi 2016 There are fundamental components which are required in OLAP and the how to implement OLAP as follows [5][6][7][10]: Database in clusing data Distictive relationship among data or tables. Dashboard to view Multidimensional data In this case multiple table, need to adjust to build necessary data component in OLAP, it is very important to include or exclude table dan relationship among tables. Correlation between table and relationship express below: Set the table (table: a collection of rows and columns). Relationships: connectedness between one table to another table. as the desired time span limitation prediction calculations. Basically, this algorithm will make reconstruction decision trees with linear regression function, for each node is formed. There are three major steps to do that [7][18][20][21]: 1. Decision-tree induction intended to build logic tree, the branches will continue to be formed, and the execution process at this stage will stop if the instance is no longer found new variations. 2. The tree which already made the first step will be cut, with regression techniques. 3. The third step, which is to avoid fluctuating conditions and avoid inconsistencies, would smoothing technique will reconstruct the logic tree form, includes branches and leaves. This is done by combining the reconstruction process with a linear model that will produce a predicted value. III. MODEL, ANALYSIS, AND DESIGN Fig. 1. Representation data cubicles in different prespective using dimension, measure and schema [11] To establish multidimensional data required a lot of tables that are related to one another, it is important in establishing the completeness of the information.in addition to need data that depends on other tables, multidimensional data is to have its own attributes to be managed in OLAP [8][9]. These three attributes are: Dimensions: attributes are reviewed. Measurement: quantity that can be measured referring to the wedge between dimensions reviewed. Calculation, which is measurement results: measurement results of a wedge (measurement). By building the OLAP data is organized in tables can be empowered for the purposes of analysis. Valuable information not found when relying on a single table only organizes rows and columns only [7][8][9]. Predictive Algorithm To extend the reliability of the OLAP Result for future used that was been produced, prediction method using M5P Algorithm will be implemented. M5P algorithm will analyze and make the value of output by values collection of instance and feature vectors, as well To form appropriate cubicle data from transactional data, the raw data must be organized in advance in realtional tables. Furthermore, the relation was made to build attachment between one table with others. Data exploration in OLAP will be implemented after the cubicles data is formed, the process of data selection such as: Summarization Collecting Large Amounts Of Data Filtering Sorting Rank (Rank) Compare multiple sets of data Sketching / charts / diagrams Findingdata pattern. Analyze data trends. Fig. 2. Cycle with OLAP data processing, mapping of data into cubes and extend the OLAP result with Prediction algorithm [10] 176

To obtain such knowledge, the way in which to exposevaluable facts from the data which is using OLAP data with filtering. OLAP able to map data in the form dimensional cube (dimensional cubes), and then each of the cubes can be easily compared, Party decision makers easily and quickly when looking for the causes of the problems faced. The software used to implement the data of restaurant transaction to perform incision and rotation data immediately to produce information that is OLAP Cube [12][13][15].This application will read the data stored in Microsoft Access data format. Furthermore, the selection of dimension and measures that would like to be processed through the build Cube by using OLAP Cube. Getting the raw data is done by taking from the sale and item data collected in 3 months data. With the characteristics of the 22 attributes and 4800 record. Attributes mean the number of items sold and the types of waste with a total value of sales, while the record refers to the number of transactions over a period of 3 months. So that data can be identified for filtering use OLAP Cube, then the data must be transformed into multiuser database which is suitable for midsize corporation. Fig. 3. Elements database relations in third normal form to be transformed into the data cubical In this case, only a portion of the data to be retrieved from the database transactional mapped in one single table after the process of de-normalize by executing the command Structured Query Language (SQL). Table 1. Results Of Normalization Which Will Be Transformed Into Data Cubicles In this point, the data can already be mapped in the OLAP cube with the formation of the third normal form, which is then performed drilling techniques with OLAP dashboard interface. After the exploration from OLAP Dashboard by using OLAP feature such as slicing, dicing, schema builder, the result will examine later using prediction algorithm to simulate whether the OLAP result describe the truth in the long run, by doing this, the decision maker can make adjustment related to the business actions which will be implemented. IV. IMPLEMENTATIONS A. Slicing and Dicing Previously said virtual cubical data can be changed in such a way to make the findings of other more valuable. This is done using a number of methods of operation of the virtual cubical data. In this last observation carried data that includes: Make an incision / Slicing. Create many slices / Dicing. Drill Up. Drill Down. Rotation or pivoting. Here will be described one by one for each operation on OLAP.Incision or slicing is taken or a onedimensional slice of virtual data for the purposes of simplification cubical information or to remove information that is not needed in the analysis. Multiple slices or wedges dicing are carried out over a twodimensional data. The dimensions of time and space limitations have not changed, but the three categories of items were taken for further analysis. B. OLAP Drill-up and Drill Down Drill Down and Drill Up is an analysis technique to specify or generalize information. Drill up (UP), the information obtained is more compact and more Drill Down (Down) then obtained more detailed information. Cubical virtual (left) if translated into cubical (right) should be referred to Drill Down from right to left is called Drill UP.Drill Down viewed as areas that fall within the product transaction such as monthly,weekly, daily or certain special occasion. C. OLAP rotation and pivoting Rotation or pivoting is done by twisting or rotating the virtual cubical data to get a different perspective on the data analyzed. Rotation cubical virtual data in such a way, will change the focus of analysis, such analysis is based on the type of food product to be based on the period. The findings of the new information could emerge as the degradation of certain insurance services annually. 177

Seminar Nasional Teknologi Informasi 2016 Before rotating, it did not see because the number of customer still has a high level of loyalty to the restaurant. The levels of use in analyzing OLAP data is relatively easy, the following steps: Connecting to OLAP data sources or data source to be analyzed. Add the necessary tables from the data source. Build relationships between tables. Selecting variables by drag and drop field or column in the table in the 'dimension' and 'measure'. Build cubical data. Any data can be used as long as it has the ODBC Driver (middleware to the database so that the database can be identified by different systems) that are suitable for the OLAP database that can identify the data.so that data can be identified for slicing attribute based on OLAP Cube, then the data in a single table must integrated on OLAP application completely. Attribute refers to the number of product. The way top executive view the data can be modified by considering product, selling product daily or even the specific period of time or in special occasion, including unsold product. Fig. 6. Different kind of perspective to view data incubicalplatform which reveal the customer choice in certain time Knowledge is revealed which can simultaneously be used as the basis for making business strategies of transactional data related to sales of food such as:shows the number of sandwich outdated and discarded every day for 3 months, showing the number of sandwich packs obsolete and discarded every day for three months, the number of muffin dumped every day for three months, the number of cakes were discarded because of unsold every day for 3 months, the amount of fruit cup that is discarded because it is not sold each day for 3 months. For future prediction, the very best of 4food product items which is exposed from OLAP will be analyzed using prediction algorithm M5P. The data set which was used related to the data considering the way customers buy 7 best product items. The result of prediction algorithm will be compared to the real transaction value in certaint periode of time. The prediction can be seen below: Table 2Analysis The Accuracy Of The Prediction For 4 Very Best Product Selling Daily Fig. 4. Items unsold in certain period of time Fig. 5. Drilling cube data for distinctive product which isunsold every day Table one showed proceeds by applying slicing and dicing OLAP method and view the seven types of products in 3 month and have an accuracy of 93% and accuracy rate degraded in the next month wich is 91%, and for the 3 month the accuracy is about to 86%. Such condition will limit the used extended OLAP Result of the next 3 month only, but still have a good result for next view months which is the rules with the accuracy above 70%. 178

V. RECOMMENDATIONS Further development according this research can be done by: - To gather more transactional data related to the datamart which should be feed from the updated data to get rich information from transactional data. - Slicing and dicing method with rich perspective can be developed to reveal the variation and richer data schema. - Multidimensional can be selected detailed related to the spesific product which will be examine for predicion. In this case prediction targeting to spesific product before used as business action in advance. - Another prediction algorithm can be used as a comparison to asses reliability of distictive prediction algorithm. VI. CONCLUSION OLAP system enables executive or decission maker to build a good analytical decision base onreal dan factual database pattern. In this case, OLAP technique which is implemented on transactional restaurant data are capable to reveal the information on the data transaction, in many perspective such as: pie charts, bar chart, line chart dan list of data, and many feature can be used to find the precious information by changing perspective using data cube. Information on the findings of OLAP restaurants can be followed up by the top executive level personnel to anticipate the needs of stocks of raw materials and space for customers in certain period of time.restaurant operators can determine the stock of each item, which is the needs of food material, or unneeded materials to minimize the number of losses, furthermore such can be determining sales strategies and trends of food. Prediction algorithm, in this case by using M5P will increase the reliability of the information for the next use. ACKNOWLEDGMENT This work is partly supported by Widyatama University. REFERENCES [1]. Agrawal, P.C., Analysis of Data Mining by Drawing OLAP Cube Models.International Journal of Advanced Research incomputer Science and Software Engineering. Volume 5, Issue 12,December 2015.ISSN: 2277 128X. [2]. Akintola K.G., Adetunmbi A.O. and Adeola O.S., Building Data Warehousing and Data Mining from Course Management Systems: A Case Study of FUTA Course Management Information Systems. International Journal of Database Theory and Application Vol. 4, No. 3, September, 2011 [3]. Amandeep Kour. Data Warehousing, Data Mining, OLAP and OLTP Technologies Are Indispensable Elements to Support Decision-Making Process in Industrial World. International Journal of Scientific and Research Publications, Volume 5, Issue 5, May 2015.ISSN 2250-3153. [4]. Bernardino J., et al. Approximate Query Answering Using Data Warehouse Striping. Journal of Intelligent. 2002 [5]. Body M., et al. A Multidimensional and Multiversion Structure for OLAP Applications DOLAP 02, November 8, McLean, Virginia, USA. 2002. [6]. Chaudruri. S., Dayal U. An Overview of Data Warehousing and OLAP technology. Sigmod Record. 1997. [7]. Chengjun Zhan, Albert Gan, and Mohammed Hadi. Prediction of Lane Clearance Time of Freeway Incidents Using the M5P Tree Algorithm. IEEE Transactions On Intelligent Transportation Systems, Vol. 12, No. 4, December 2011.And Technology, Vol. 4, No. 1,February 2013. [8]. Codd, E.F., et al. C.T.: Providing OLAP to User Analysts: an it Mandate. Technical Report, E.F. Codd &Associates. 1993. [9]. D. Burdick, P. M. Deshpande, T. S. Jayram, R. Ramakrishnan, and S. Vaithyanathan. OLAP over uncertain and imprecise data. In VLDB 2005. [10]. Feri Sulianta. OLAP Excel Cara Hebat Excel Mengelola Data. Elexmedia Komputindo. Jakarta. 2011. [11]. Feri Sulianta. Teknik Modern Perancangan Arsitektur Sistem Informasi. Penerbit Andi. Yogyakarta. 2016. [12]. H.-J. Lenz and B. Thalheim. OLAP Databases and Aggregation Functions. In SSDBM 2001. [13]. I Dewa Made Adi Baskara Joni, Muhamad Nurudin. Penerapan Olap Untuk Monitoring Kinerja Perusahaan. Seminar Nasional Sistem Informasi Indonesia, 2-4 Desember 2013. 2013 Sesindo. [14]. Rian Pratama. Perancangan Data Warehouse Pemetaan Data SiswaPada Disdikpora Kota Palembang. STMIK GI MDP. [15]. Rui Oliveira, Jorge Bernardino. Building olap tools over large databases. Isec Instituto Superior de Engenharia de Coimbra, Polytechnic Institute of Coimbra. Quinta da Nora, Rua Pedro Nunes, P-3030-199 Coimbra, Portugal. [16]. S. Abiteboul, R. Hull, and V. Vianu. Foundations of Databases. Addison-Wesley, 1995. [17]. S. I. McClean, B. W. Scotney, and M. Shapcott. Aggregation of Imprecise and Uncertain Information in Databases. IEEE TKDE, 13(6):902 912, 2001. [18]. Susanto, S.Kom, M.Kom. Penerapan Data Mining Classification Untuk Prediksi. Perilaku Pola Pembelian Terhadap Waktu Transaksi Menggunakan Metode Naïve Bayes. Konferensi Nasional Sistem & Informatika 2015.STMIK STIKOM Bali, 9-10 Oktober 2015. [19]. T. B. Pedersen, C. S. Jensen, and C. E. Dyreson. Supporting Imprecision in Multidimensional Databases Using Granularities. In SSDBM, 1999. [20]. Wang,Y., Witten, I. H.: Induction of model trees for predicting continuous classes. In: Poster papers of the 9th European Conference on Machine Learning, 1997. [21]. Witten,Ian., Frank, Eibe., Data Mining Practical Machine Learning Tool and techniques. Morgan Kaufmann Publishers 2005 page :6 ; page: 27 paragraph 1 ; page : 112-118, page 47-86. 179