Chapter 11 Review Questions
Define and describe robotics and give examples of their application in organization.
- Deals with the design, construction, operation and use of movable machines that can substitute for humans along with computer systems for their control, sensory feedback and information processing. - Manufacturing robotics is the most common such as Renault SA plant that uses them to drive screws into engines.
Describe how the following systems support knowledge work: CAD, virtual reality, and augmented reality
CAD: - automates the creation and revision of designs, using computers and sophisticated graphics software - Allows the designer to use simulations of a physical prototype before the actual printing of the design. - They also supply data for 3-D printing for physical prototyping Virtual Reality Visualization, rendering and simulation capabilities to go far beyond CAD systems Augmented Reality Enhances visualization by overlaying digital data and images onto a physical real-world environment. Provides additional information to enhance the perception of reality so that it's more interactive and meaningful. Example is the yellow first-down markers shown on televised football games.
Distinguish between data, knowledge and wisdom and between tacit knowledge and explicit knowledge.
Data: flow of events or transactions captured by the org's systems. They are useful for transactions in this state. Information: organized data into categories of understanding. Knowledge: discover patterns, rules and contexts where the knowledge works Wisdom: collective and individual experience of applying knowledge to the solution of problems. Tactic Knowledge: knowledge that is stored in the brains of employees, not documented. Explicit Knowledge: documented knowledge
Define and describe the various types of enterprise-wide knowledge management systems and explain how they provide value for businesses.
Enterprise Content Management Systems - Help organizations manage structured (explicit knowledge in formal documents and formal rules) and semistrucutred knowledge assets (folders, messages, memos, proposals, emails, graphics, presentations, videos). Collaboration and Social Tools Learning Management Systems
Define artificial intelligence and the major AI techniques
AI in an ambitious definition is the attempt to build computer systems that act, and think like humans, but in the raw form of the definition, AI takes data input from the environment, processes the data and produces output. Where they differ is in the techniques and tech that they use to perform these activities. Driving forces for AI include: - Development of big data by the internet, e-commerce IoT and social media - Drastic reduction in cost of processing power and growth in the power of the processors - Refinement of algorithms and significant investment from business and governments Major Techniques - Expert Systems - Machine Learning - Neural networks and deep learning - Genetic algorithms - Natural language processing - Computer vision systems - Robotics - Intelligent agents
Describe the stages in the knowledge management value chain
Data and Information Acquisition: - collecting, storing, disseminating (based on feedback) Acquire - IS:Business analytics, Data mining, neural networks, machine learning, knowledge workstations, expert knowledge networks - Mgmt & Org: knowledge culture, CoP, social networks Store - IS: Content Management Systes, Knowledge databases, expert systems - Mgmt & Org: Org routines, org cultures Disseminate - IS: Portals, Search Engines, Collaboration and social business tools - Mgmt & Org: Training, Collaboration Apply - IS: DSS, Enterprise Applications, Robotics - Mgmt & Org: New IT-based business processes, New products and services, new markets
What are the 3 types of knowledge management systems?
Enterprise-wide KM systems - General purpose, integrated firmwide efforts to collect, store, disseminate and use digital content and knowledge - Enterprise content management systems - Collaboration and social tools - Learning management systems Knowledge Work Systems - Specialized workstations and systems that enable scientists, engineers and other knowledge workers to create and discover new knowledge. - Computer-aided design (CAD), Virtual reality Intelligent Techniques - They consist of tools for discovering patterns and applying knowledge to discrete decisions and knowledge domains. - Data mining, expert systems, machine learning, neural networks, natural language processing, computer vision systems, robotics, genetic algorithms and intelligent agents.
Define knowledge work systems and describe the generic requirements of knowledge work systems.
Knowledge workers have 3 key roles: - Keep the org current in knowledge as it develops in the external world - Serving as internal consultants in their area of knowledge, changes taking place and opportunities - Acting as change agents, evaluating, initiating and promoting change projects Knowledge Work Systems Requirements - Powerful graphics - Analytical tools - Communications and document management capabilities - Sufficient computing power to handle sophisticated graphics or complex calculations - Provide quick and easy access to external databases User-friendly interfaces
Define and describe Natural Language Processing and give examples of their application in organization.
Natural Language Processing - Makes it possible for a computer to understand and analyze natural language, not specifically formatted to be understood by computers. - Based on machine learning and often uses deep learning. Examples: Akershus Hospital used LP and IBM Watson Explorer to sift through medical records with unstructured text data expressed in NL. It used the algorithms to read the medical record and determine its meaning. Mizuho Bank used speech recognition and IBM Watson content analytics to convert customer speech to text data and apply the NLP to learn more from each customer interaction so that it can eventually infer the customer's need and display that to the agent.
Define machine learning, explain how it works and give examples of the kinds of problems it can solve.
Software that can identify patterns in very large databases without explicit programming although with significant human training. How it Works - Find patterns in data and classifying those data inputs into known and unknown outputs - Instead of expert knowledge, it uses very large data sets to automatically find patterns and relationships, analyzing large set of examples and making statistical inferences. - 2 different types of Learning include supervised learning in which the machine is fed a large amount of data and then humans will "show" the machine how to identify a particular type of data within the set. The machine can then learn from being shown the different types of particular data on its own. This has a very high accuracy rating. The other type is unsupervised where the machine is given a large amount of data, but instead of being shown how to discover the pattern, it's given an insanely large amount of processing power (brute force) and will eventually learn and figure it out on it's own. This has a very low rate of accuracy, but is continuing to improve over time. Benefits - Extraordinary ability to recognize patterns at scale in extremely shortened time periods - Ability to classify objects into discrete categories Limitations - It needs to have a very large datasets and computing facilities - The most desired outcomes need to already be defined by humans - Output is binary - The need for talented and large groups of software/system engineers is required. Examples: facebook ad display which uses prior user behavior information, information supplied by advertisers, user activity on apps and other websites that facebook can track. It uses that data to estimate within seconds the probability that any specific user will actually click on the add.
Describe the roles of the following in facilitating knowledge management: taxonomies, MOOCs, and learning management systems.
Taxonomies: a classification scheme meant to organize the information into meaningful categories so it can easily be accessed. Learning Management Systems: provides tools for management, delivery, tracking and assessment of vaious types of employee learning and training. MOOCs: massive open online courses used to educate employees through an online course made available via the web to a very large number of participants. LinkedIn Learning probably.
Define and describe genetic algorithms and intelligent agents. Explain how each works and the kinds of problems for which each is suited.
Genetic Algorithms: useful for finding optimal solutions for problems by examining a very large number of alternative solutions for the problem. How it works - Searches a population of randomly generated strings of binary digits to identify the right string representing the best possible solution for the problem. - As the solutions alter and combine, the worst ones are discarded and the better ones go on to produce better solutions. Benefits - They can solve dynamic and complex problems that involve hundreds or thousands of variables or formulas. - The problem must be one whose range of possible solutions can be represented genetically and criteria can be established for evaluating fitness Examples: General Electric uses this for optimizing the design of jet turbine aircraft engine since it can require changes in up to 100 variables for just one change. Intelligent Agents: software programs that work in the background without direct human intervention to carry out specific tasks for an individual user, business process, or software application. Examples: Siri which over time can also learn the user preferences; chatbots for automated conversations such as Vodafone who uses it to answer questions; simulations
Define and describe computer vision systems and give examples of their application in organization.
How computers can emulate the huan visual system to view and extract information from real-word images. Facebook uses deepFace to recognize faces to ensure that every photo in facebook is associated with a profile.
Describe the important dimensions of knowledge
Knowledge is a Firm Asset - Useful information requires data transformation and knowledge requires org resources - Knowledge is an intangible asset - Knowledge is not susceptible to the law of diminishing returns and instead gains value the more people share it. Knowledge has Different Forms - Tactic or Explicit - Involves know-how, craft and skill - Involves knowing how to follow procedures - Involves knowing why, and when, things happen. Knowledge has Location - A cognitive event involving mental models and maps of individuals - Social and individual basis of knowledge - It's "sticky", situated, and contextual Knowledge is situational - It's conditional, knowing when to apply a procedure is just as important as knowing the procedure. - Related to context; you need to know how to use the tool and under what circumstances.
Define neural networks and deep learning neural networks, describing how they work and how they benefit organizations
Neural network is composed of interconnected units called neurons that can take and transfer data to/from other neurons. The neurons are software programs and mathematical models that perform the input/output functions. The weight of the connections is controlled by a Learning Rule, an algorithm that systematically alters the strength of the connections to produce the final desired output. Neural networks: find patterns and relationships in very large amounts of data that is too complicated and difficult for human to analyze by machine learning algorithms and computational models loosely based on how the biological human brain operates. They are pattern detection programs. Deep Learning: more complex with many layers of transformation of the input data to produce the target output. It's expected to be "self-taught" and looks for patterns in the data without specific direction of what to look for. How it works - They learn patterns by sifting through large quantities of data and finding pathways through the networks. - Once a successful path is found, the Learning Rule is used to strengthen that connection between the neurons in the pathways. - The process is repeated thousands/millions of times until the most successful pathways are discovered. Eventually the process will stop when an acceptable level of pattern recognition is reached. Limitations - Require very large data sets - Often many patterns in large data sets are nonsensical and requires humans to choose which patterns make sense. - It's hard to explain how the system arrived at the solution - They have no understanding of ethics Examples: - Computer vision used for photo tagging, facial recognition, autonomous vehicles - Speech recognition used for digital assistants, chatbots, helpcenters - Machine controls, diagnostics: preventive maintenance; quality control - Language translation: translate sentences from one language to another - Transaction analysis used for fraud control; theft of services; stock market predictions - Targeted online ads: programmatic advertising
Define knowledge management and explain its value to businesses.
Note: 20% of total economic output of the US (4 trillion) is derived from output of the information and knowledge sectors of business. Organizational Learning: Through the collection of data, measurements, experimentation and feedback from customers, organizations gain experience. From this experience, organizational learning occurs as a process by which organizations change to adjust their behavior to reflect the learning by creating new business processes and by changing patterns of management decision making. Knowledge Management: the set of business processes developed in an organization to create, store, transfer and apply knowledge. Value: it increases the ability of the organization to learn from its environment and incorporate knowledge into its business process.
Define an expert system, describe how it works and explain its value to business.
Represent knowledge of experts in an if-then set of rules and used to assist in decisions. How it Works: - Uses a knowledge base: collective set of rules - Inference Engine: strategy used to search through the collection of rules and formulate the conclusions which works because it is triggered by facts the user gathers and enters. Benefits - Improved decision making - Reduced errors - Reduced costs - Reduced training time - Better quality and service Limitations - Sometimes experts can't say HOW they make a decision - Rapidly changing rules causes too much overhead for keeping the system up-to-date - Do not handle unstructured decision making which is the most common type - Do not scale well - Expensive to build Examples: application for making a decision on granting credit or medical diagnostics.
