“What is the fastest way to move from one point to another, achieving your goals? Without software tools and an understanding of data, you will be groping in the dark.”
“With Business Intelligence, you get an insight to every business process, with this knowledge, there is no need for destructive competition, solving problems for maximum ROI.”
– Contributed by Oogle.
Applied Multiple Regression Analysis by filling in the missing blanks ;
Two Way Anova
Most internal auditors, especially those working in customer-focused industries, are aware of data mining and what it can do for an organization — reduce the cost of acquiring new customers and improve the sales rate of new products and services. However, whether you are a beginner internal auditor or a seasoned veteran looking for a refresher, gaining a clear understanding of what data mining does and the different data mining tools and techniques available for use can improve audit activities and business operations across the board.
WHAT IS DATA MINING?
In its simplest form, data mining automates the detection of relevant patterns in a database, using defined approaches and algorithms to look into current and historical data that can then be analyzed to predict future trends. Because data mining tools predict future trends and behaviors by reading through databases for hidden patterns, they allow organizations to make proactive, knowledge-driven decisions and answer questions that were previously too time-consuming to resolve.
Data mining is not particularly new — statisticians have used similar manual approaches to review data and provide business projections for many years. Changes in data mining techniques, however, have enabled organizations to collect, analyze, and access data in new ways. The first change occurred in the area of basic data collection. Before companies made the transition from ledgers and other paper-based records to computer-based systems, managers had to wait for staff to put the pieces together to know how well the business was performing or how current performance periods compared with previous periods. As companies started collecting and saving basic data in computers, they were able to start answering detailed questions quicker and with more ease.
Changes in data access — where there has been greater empowerment and integration, particularly over the past 30 years — also have impacted data mining techniques. The introduction of microcomputers and networks, and the evolution of middleware, protocols, and other methodologies that enable data to be moved seamlessly among programs and other machines, allowed companies to link certain data questions together. The development of data warehousing and decision support systems, for instance, has enabled companies to extend queries from “What was the total number of sales in New South Wales last April?” to “What is likely to happen to sales in Sydney next month, and why?”
However, the major difference between previous and current data mining efforts is that organizations now have more information at their disposal. Given the vast amounts of information that companies collect, it is not uncommon for them to use data mining programs that investigate data trends and process large volumes of data quickly. Users can determine the outcome of the data analysis by the parameters they chose, thus providing additional value to business strategies and initiatives. It is important to note that without these parameters, the data mining program will generate all permutations or combinations irrespective of their relevance.
Internal auditors need to pay attention to this last point: Because data mining programs lack the human intuition to recognize the difference between a relevant and an irrelevant data correlation, users need to review the results of mining exercises to ensure results provide needed information. For example, knowing that people who default on loans usually give a false address might be relevant, whereas knowing they have blue eyes might be irrelevant. Auditors, therefore, should monitor whether sensible and rational decisions are made on the basis of data mining exercises, especially where the results of such exercises are used as input for other processes or systems.
Auditors also need to consider the different security aspects of data mining programs and processes. A data mining exercise might reveal important customer information that could be exploited by an outsider who hacks into the rival organization’s computer system and uses a data mining tool on captured information.
DATA MINING TOOLS
Organizations that wish to use data mining tools can purchase mining programs designed for existing software and hardware platforms, which can be integrated into new products and systems as they are brought online, or they can build their own custom mining solution. For instance, feeding the output of a data mining exercise into another computer system, such as a neural network, is quite common and can give the mined data more value. This is because the data mining tool gathers the data, while the second program (e.g., the neural network) makes decisions based on the data collected.
Different types of data mining tools are available in the marketplace, each with their own strengths and weaknesses. Internal auditors need to be aware of the different kinds of data mining tools available and recommend the purchase of a tool that matches the organization’s current detective needs. This should be considered as early as possible in the project’s lifecycle, perhaps even in the feasibility study.
Most data mining tools can be classified into one of three categories: traditional data mining tools, dashboards, and text-mining tools. Below is a description of each.
- Traditional Data Mining Tools. Traditional data mining programs help companies establish data patterns and trends by using a number of complex algorithms and techniques. Some of these tools are installed on the desktop to monitor the data and highlight trends and others capture information residing outside a database. The majority are available in both Windows and UNIX versions, although some specialize in one operating system only. In addition, while some may concentrate on one database type, most will be able to handle any data using online analytical processing or a similar technology.
- Dashboards. Installed in computers to monitor information in a database, dashboards reflect data changes and updates onscreen — often in the form of a chart or table — enabling the user to see how the business is performing. Historical data also can be referenced, enabling the user to see where things have changed (e.g., increase in sales from the same period last year). This functionality makes dashboards easy to use and particularly appealing to managers who wish to have an overview of the company’s performance.
- Text-mining Tools. The third type of data mining tool sometimes is called a text-mining tool because of its ability to mine data from different kinds of text — from Microsoft Word and Acrobat PDF documents to simple text files, for example. These tools scan content and convert the selected data into a format that is compatible with the tool’s database, thus providing users with an easy and convenient way of accessing data without the need to open different applications. Scanned content can be unstructured (i.e., information is scattered almost randomly across the document, including e-mails, Internet pages, audio and video data) or structured (i.e., the data’s form and purpose is known, such as content found in a database). Capturing these inputs can provide organizations with a wealth of information that can be mined to discover trends, concepts, and attitudes.
Besides these tools, other applications and programs may be used for data mining purposes. For instance, audit interrogation tools can be used to highlight fraud, data anomalies, and patterns. An example of this has been published by the United Kingdom’s Treasury office in the 2002–2003 Fraud Report: Anti-fraud Advice and Guidance, which discusses how to discover fraud using an audit interrogation tool. Additional examples of using audit interrogation tools to identify fraud are found in David G. Coderre’s 1999 book, Fraud Detection.
In addition, internal auditors can use spreadsheets to undertake simple data mining exercises or to produce summary tables. Some of the desktop, notebook, and server computers that run operating systems such as Windows, Linux, and Macintosh can be imported directly into Microsoft Excel. Using pivotal tables in the spreadsheet, auditors can review complex data in a simplified format and drill down where necessary to find the underlining assumptions or information.
When evaluating data mining strategies, companies may decide to acquire several tools for specific purposes, rather than purchasing one tool that meets all needs. Although acquiring several tools is not a mainstream approach, a company may choose to do so if, for example, it installs a dashboard to keep managers informed on business matters, a full data-mining suite to capture and build data for its marketing and sales arms, and an interrogation tool so auditors can identify fraud activity.
DATA MINING TECHNIQUES AND THEIR APPLICATION
In addition to using a particular data mining tool, internal auditors can choose from a variety of data mining techniques. The most commonly used techniques include artificial neural networks, decision trees, and the nearest-neighbor method. Each of these techniques analyzes data in different ways:
- Artificial neural networks are non-linear, predictive models that learn through training. Although they are powerful predictive modeling techniques, some of the power comes at the expense of ease of use and deployment. One area where auditors can easily use them is when reviewing records to identify fraud and fraud-like actions. Because of their complexity, they are better employed in situations where they can be used and reused, such as reviewing credit card transactions every month to check for anomalies.
- Decision trees are tree-shaped structures that represent decision sets. These decisions generate rules, which then are used to classify data. Decision trees are the favored technique for building understandable models. Auditors can use them to assess, for example, whether the organization is using an appropriate cost-effective marketing strategy that is based on the assigned value of the customer, such as profit.
- The nearest-neighbor method classifies dataset records based on similar data in a historical dataset. Auditors can use this approach to define a document that is interesting to them and ask the system to search for similar items.
Each of these approaches brings different advantages and disadvantages that need to be considered prior to their use. Neural networks, which are difficult to implement, require all input and resultant output to be expressed numerically, thus needing some sort of interpretation depending on the nature of the data-mining exercise. The decision tree technique is the most commonly used methodology, because it is simple and straightforward to implement. Finally, the nearest-neighbor method relies more on linking similar items and, therefore, works better for extrapolation rather than predictive enquiries.
A good way to apply advanced data mining techniques is to have a flexible and interactive data mining tool that is fully integrated with a database or data warehouse. Using a tool that operates outside of the database or data warehouse is not as efficient. Using such a tool will involve extra steps to extract, import, and analyze the data. When a data mining tool is integrated with the data warehouse, it simplifies the application and implementation of mining results. Furthermore, as the warehouse grows with new decisions and results, the organization can mine best practices continually and apply them to future decisions.
Regardless of the technique used, the real value behind data mining is modeling — the process of building a model based on user-specified criteria from already captured data. Once a model is built, it can be used in similar situations where an answer is not known. For example, an organization looking to acquire new customers can create a model of its ideal customer that is based on existing data captured from people who previously purchased the product. The model then is used to query data on prospective customers to see if they match the profile. Modeling also can be used in audit departments to predict the number of auditors required to undertake an audit plan based on previous attempts and similar work.
Using data mining to understand and extrapolate data and information can reduce the chances of fraud, improve audit reactions to potential business changes, and ensure that risks are managed in a more timely and proactive fashion. Auditors also can use data mining tools to model “what-if” situations and demonstrate real and probable effects to management, such as combining real-world and business information to show the effects of a security breach and the impact of losing a key customer. If data mining can be used by one part of the organization to influence business direction for profit, why can’t internal auditors use the same tools and techniques to reduce risks and increase audit benefits?
John Silltow has more than 20 years’ experience working with government and financial information systems in England, focusing on computer audit and security. He is now managing director of his own company, Security Control and Audit Ltd., and specializes in Internet security, software management, and IT and audit training.
– Contributed by Oogle.