MIS chapter 13

Contextual Computing

: Making Mobile Devices Smarter •Computing environment that is always present •Capable of offering recommendations based on who we are, where we are, and whom we are with •New app in development adds capability to perceive a user's mood through analysis of phone calls, text messages, and internet access •Based on the principle that computers can both sense and react to their environments

Interfacing

Accepting human language as input •Carrying out the corresponding command •Generating the necessary output

Web marketing

Collect information about customers and use it to better market products and services

Factual knowledge

Facts related to a specific discipline, subject, or problem. For example, facts related to kidney problems might include kidney size, blood levels of certain enzymes, and duration and location of pain.

Shopping and information agents

Help users navigate through vast resources available on the Web and provide better results in finding information •Serve as search engines, site reminders, or personal surfing assistants

Components of an Expert System

Knowledge acquisition facility, knowledge base, knowledge base management system, user interface, explanation facility, inference engine

Uses of Expert Systems

Many companies are engaged in research and development of expert systems, and these systems are now used in areas such as the following: •Airline industry •Forensics lab work •Banking and finance •Education •Agriculture and food industry •Personnel management •Security and U.S. government

Meta-knowledge

Meta-knowledge is knowledge about knowledge. It enables an expert system to learn from experience and examine and extract relevant facts to determine the path to a solution. It also guides future planning or execution phases of an expert system. For example, knowing how an expert system makes decisions is considered meta-knowledge. Although this type of knowledge is not currently available in expert systems, integrating neural networks into expert systems is one possibility for acquiring meta-knowledge.

Personal agents

Perform specific tasks for a user, such as remembering information for filling out Web forms

Heuristic knowledge

Rules related to a problem or discipline. For example, the general rules indicating that a patient has a kidney problem could include severe pain in the lower left or lower right of the back and high levels of creatinine and blood urea nitrogen.

Virtual catalogs

Smart or interactive catalogs •Display product descriptions based on customers' previous experiences and preferences

Monitoring and surveillance agents

Track and report on computer equipment and network systems to predict when a system crash or failure might occur

Knowledge acquisition

Using the computer to read large amounts of text, understand the information, and summarize important points

AI used in two types of decision-making analyses

What-is analysis, What-if analysis •AI helps to answer questions such as why, what it means, what should be done, and when should it be done

Data mining agents

Work with a data warehouse •Detect trends and discover information and relationships among data items that were not readily apparent

Augmented intelligence

complements a decision maker capabilities, not replaces him or her. unlike AI that replaces him or her. •The goal is to help the decision maker work smarter and faster •Innovations in AI chips result in more smart devices, increasing efficiency and effectiveness •Augmented intelligence benefits diverse industries including manufacturing, health care, construction, logistics and distribution, automotive and transportation, and security.

Heuristic data

encourages applying knowledge based on experience to find a solution to a problem

Forward chaining

series of "if-then-else" condition pairs is performed The "if" condition is evaluated first, then the corresponding "then-else" action is carried out. For example, "if" the temperature is less than 80F° and the grass is 3 inches long, "then" cut the grass or "else" wait. In a medical diagnostic expert system, the system could evaluate a problem as follows: If the patient's temperature is over 101F° and if the patient has a headache, then it's very likely (a 95 percent chance) that the patient has the flu, or else search for other diseases.

Backward chaining

starts with the goal and backtracks to find the right solution In other words, to achieve this goal, what conditions must be met? To understand the differences between these two techniques, consider the following example. In an expert system that provides financial investment advice for investors, the system might use forward chaining and ask 50 questions to determine which of five investment categories—oil-gas, bonds, common stocks, public utilities, and transportation—is more suitable for an investor

What-if analysis

used in decision support systems. Decision makers use it to monitor the effect of a change in one or more variables. It is available in spreadsheet programs, such as Microsoft Excel.

What-is analysis

used in transaction-processing systems and management information systems. For example, if you enter a customer account number, the system displays the customer's current balance. However, these systems lack the capability to report real-time information or predict what could happen in the future. For example, reports generated by accounting information systems that show performance over the preceding fiscal quarter consist of past events, so decision makers cannot do much with this information.

Integrating AI Technologies into Decision Support Systems

•AI-related technologies can improve the quality of decision support systems (DSSs) •Result in integrated or intelligent DSSs (IDSSs) •Add explanation capabilities by integrating expert systems •Add learning capabilities by integrating ANNs •Create a user-friendly interface by integrating an NLP system •Benefits of integrating expert systems into the database component of a DSS •Adding deductive reasoning to traditional DBMS functions •Improving access speed and database creation and maintenance •Adding capability to handle uncertainty and fuzzy data •Simplifying query operations

Fuzzy Logic

•Allows a smooth, gradual transition between human and computer vocabularies •Deals with variations in linguistic terms by using a degree of membership in a set •Designed to help computers simulate vagueness and uncertainty in common situations •Allows computers to reason similarly to humans •Used in several areas •Search engines •Chip design •Database management systems •Software development •Appliances

Ethical Issues of AI

•Before implementing an AI system, organizations should establish an ethical framework •Define the AI goals •Define the complexity of the problem •Define the environment as being stable or variable •Define and guard against bias •Define the level of human involvement •Widespread adoption of AI technology raises major issues, including: •AI bias •AI mistakes •Wealth inequality •Impact on humanity and human behavior •Impact on unemployment

Natural-Language Processing

•Developed so that users can communicate with computers in human language •Provides a question-and-answer setting that is natural and easier for people to use •Useful with databases •Disadvantage •Complexity of the human language renders the development of NLP systems difficult •Categories •Interface to databases •Machine translation •Text scanning and intelligent indexing programs •Generating text for automated production of standard documents •Speech systems for voice interaction with computers •Activities performed •Interfacing •Accepting human language as input •Carrying out the corresponding command •Generating the necessary output •Knowledge acquisition •Using the computer to read large amounts of text, understand the information, and summarize important points

Explanation facility

•Explains to end users how recommendations are derived performs tasks similar to what a human expert does by explaining to end users how recommendations are derived. For example, in a loan evaluation expert system, the explanation facility states why an applicant was approved or rejected. In a medical expert system, it explains why the system concluded that a patient has a kidney stone, for instance. This component is important because it helps give users confidence in the system's results.

Criteria for Using Expert Systems

•Extensive human expertise is needed but a single human cannot tackle problem alone •Necessary expertise can be represented as rules or heuristics •Decision or task has already been handled successfully by human experts •Requires consistency and standardization •Involves many rules and complex logic •Subject domain is limited •Experts in the organization are scarce

Genetic Algorithms

•Form of AI mainly used to find solutions to optimization and search problems •Algorithms that mimic the process of natural evolution •Generate solutions to optimization •Find the combination of inputs that generates the most desirable outputs •Examine search problems using mutation, selection, crossover, and chromosome techniques

Artificial neural networks (ANNs)

•Learn and are capable of performing tasks difficult with conventional computers •Used for poorly structured problems •Cannot supply an explanation for the solution •Use patterns instead of the if-then-else rules used by expert systems •Create a model based on input and output •Tasks that involve the use of ANNs •Bankruptcy prediction •Credit rating •Investment analysis •Oil and gas exploration •Target marketing •Computer and network security

Robots

•Most successful application of AI •Excel at performing simple, repetitive tasks •Free workers from tedious or hazardous jobs •Typically have limited mobility and some have limited vision •Example of a robot with advanced mobility and vision: Honda's ASIMO •Controlled by a computer program that includes commands •Different programming languages: •Variable Assembly Language (VAL) •Functional Robotics (FROB) •A Manufacturing Language (AML) •Personal robots have limited mobility, vision, and speech capabilities •Often used as prototypes to test certain services •Advantages of robots in the workplace •No interpersonal or personnel issues •Consistency •Can function in dangerous environments

Advantages of Expert Systems

•Never become distracted, forgetful, or tired •Duplicate and preserve expertise of scarce experts •Preserve expertise of employees who are retiring or leaving •Create consistency in decision making •Improve decision-making skills of nonexperts

Criteria for Not Using Expert Systems

•Presence of too few or too many rules •Involves well-structured numerical problems that don't require an expert system •Involves a broad range of topics but not many rules •Disagreement among experts •Requires human expertise

Machine Learning

•Process and procedure by which knowledge is gained through experience •Several applications •Social media and identifying faces in photos •Recognizing commands spoken into smartphones •Designing intelligent robots •Artificial neural networks (ANNs) •Learn and are capable of performing tasks difficult with conventional computers •Used for poorly structured problems •Cannot supply an explanation for the solution •Use patterns instead of the if-then-else rules used by expert systems •Create a model based on input and output

Expert Systems

•Programs that mimic human expertise in a specific area that human experts have solved successfully •To be successful, must be applied to tasks that human experts have already handled •Tasks in medicine, geology, education, and oil exploration •Also used in search engines to better understand users' queries •Work with heuristic data •Heuristic data encourages applying knowledge based on experience to find a solution to a problem

User interface

•Provides user-friendly access to the expert system This is the same as the user interface component of a decision support system. It provides user-friendly access to the expert system. Although graphical user interfaces (GUIs) have improved this component, one goal of AI technology is to provide a natural language (discussed later in the module) for the user interface.

Knowledge base management system (KBMS)

•Similar to a DBMS •Used to keep the knowledge base updated, with changes to facts, figures, and rules

knowledge base

•Similar to a database •In addition to storing facts and figures, keeps track of associated rules and explanations •Factual knowledge •Heuristic knowledge •Meta-knowledge For example, a financial expert system's knowledge base might keep track of all figures constituting current assets, including cash, deposits, and accounts receivable. It might also keep track of the fact that current assets can be converted to cash within one year. An expert system in an academic environment might include facts about all graduate students, such as GMAT scores and grade point averages (GPAs), as well as a rule specifying that classified graduate students must have a GMAT of 650 or better and a GPA of 3.4 or better.

Inference engine

•Similar to the model base component •Uses techniques such as forward and backward chaining to manipulate a series of rules •Forward chaining: series of "if-then-else" condition pairs is performed •Backward chaining: starts with the goal and backtracks to find the right solution Some inference engines work from a matrix of facts that includes several rows of conditions and rules, similar to a decision table. In this case, rules are evaluated one at a time, then advice is provided. Some inference engines also learn from doing.

Intelligent Agents

•Software capable of reasoning and following rule-based processes •Popular in e-commerce •Other names •Bots •Virtual agents (VAs) •Intelligent virtual agents (IVAs) •Characteristics: •Adaptability •Autonomy •Collaborative behavior •Humanlike interface •Mobility •Reactivity •Applications: •Web marketing •Collect information about customers and use it to better market products and services •Virtual catalogs •Smart or interactive catalogs •Display product descriptions based on customers' previous experiences and preferences •Shopping and information agents •Help users navigate through vast resources available on the Web and provide better results in finding information •Serve as search engines, site reminders, or personal surfing assistants •Personal agents •Perform specific tasks for a user, such as remembering information for filling out Web forms •Data mining agents •Work with a data warehouse •Detect trends and discover information and relationships among data items that were not readily apparent •Monitoring and surveillance agents •Track and report on computer equipment and network systems to predict when a system crash or failure might occur

Knowledge acquisition facility

•Software package with manual or automated methods for acquiring and incorporating new rules and facts •Enables growth of an expert system This component works with the knowledge base management system (described later in this list) to ensure that the knowledge base is as up to date as possible.

Artificial Intelligence

•Technologies that try to simulate and reproduce human thought behavior, including thinking, speaking, feeling, and reasoning. •Artificial intelligence (AI) technologies •Apply computers to areas that require knowledge, perception, reasoning, understanding, and cognitive abilities •Concerned with generating and displaying knowledge and facts To achieve these capabilities, computers must be able to do the following: Understand facts and manipulate qualitative data. Deal with exceptions and discontinuity. Understand relationships between facts. Interact with humans in their own language. Deal with new situations based on previous learning. Information systems are concerned with capturing, storing, retrieving, and working with data, but AI technologies are concerned with generating and displaying knowledge and facts. In the information systems field, as you have learned, programmers and systems analysts design systems that help decision makers by providing timely, relevant, accurate, and integrated information. In the AI field, knowledge engineers try to discover "rules of thumb" that enable computers to perform tasks usually handled by humans. Rules used in the AI field come from a diverse group of experts in areas such as mathematics, psychology, economics, anthropology, medicine, engineering, and physics. AI encompasses several related technologies discussed in this module, including robotics, expert systems, fuzzy logic systems, intelligent agents, artificial neural networks, genetic algorithms, and natural-language processing.

Case-Based Reasoning (CBR)

•a problem-solving technique •Matches a new case with a previously solved case and its solution •Both stored in a database •Offers a solution after searching for a match •A human expert is required to solve the problem if CBR fails to find a match If there is no exact match between the new case and cases stored in the database, the system can query the user for clarification or more information. After finding a match, the CBR system offers a solution; if no match is found, even after supplying more information, the human expert must solve the problem. •Design and implementation involves four Rs •Retrieve •Reuse •Revise •Retain

Soft robots

•simpler to make and cost less •Typical applications •High-speed food handling •Precise pick and place •Adaptive grasping •Warehouse logistics •Advanced assembly •Medical field