Download Tech_Trends_Knowledge Based_Crowe_FINAL

RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
Is Knowledge Based Artificial Augmentation Intelligence Technology the Next Step in
Instructional Tools for Distance Learning?
Dale Crowe
University of Phoenix
School of Advanced Studies
[email protected]
Cell: 541-971-3259
1109 Kamerin Springs Dr., Talent, OR 97540
Martin LaPierre
University of Phoenix
School of Advanced Studies
[email protected]
Cell: 909-210-2334
Address: 40998 Promenade Chardonnay, Temecula CA 92591
Mansureh Kebritchi
University of Phoenix
School of Advanced Studies
[email protected]
Cell: 407-496-1224
Address: 3935 Flowering Stream Way, Oviedo, FL 32766
Keywords: Artificial augmentation; Artificial intelligence; Cognitive computing, Knowledge
based systems; Machine learning; Natural language processing; Distance learning and artificial
augmentation applications; IBM Watson and distance learning tools.
1
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
2
Abstract
With knowledge based systems (KBS) it is now possible to develop distance learning
applications to support both curriculum and administrative tasks. We are now moving from the
programmable systems era initiated in the 1950s to the cognitive computing era. A KBS is
cognitive computing that understands natural language, adapts and learns, and generates and
evaluates hypotheses. The system can manage data and create tools to generate meaningful data.
As a proof of the concept, the authors conducted a feasibility study and development plan with
the input of 20 subject matter experts (programmers, instructional designers and content experts)
for a prototype KBS application for distance/online learning. Results include discussions of the
philosophical differences between artificial intelligence and the augmented intelligence approach,
augmented intelligence applications, and the role of instructional designers to support adoption
of the applications.
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
3
Introduction
Various advances in applications of augmented intelligence in distance learning have
developed, but resistance persists among educators to adopt augmented intelligence applications.
This resistance is partly due to confusion between the concept of augmented intelligence and
artificial intelligence. John McCarthy was credited with devising the term artificial intelligence
in 1956, and the philosophical conflict between these two concepts has continued for more than
half a century (Lavenda, 2016). Augmented intelligence refers to increasing the capability of
human beings to solve complex problems and gain better and faster comprehension of the
situation (Engelbart, 1962). Augmented intelligence is related to the concept of an intelligence
amplifier that Ashby (1956, 1960) first used.
Human intelligence is amplified by organizing intellectual capabilities into higher levels
of synergistic structuring with the assistance of technology (Engelbart, 1962). In opposition,
artificial intelligence refers to the capability of a machine to imitate intelligent human behavior
(Merriam Webster, 1999). Artificial intelligence is in distinct contrast to augmented intelligence.
Augmented intelligence seeks to empower and amplify human capability to solve complex
problems. It is used as a supplemental tool to support the human being. The purpose of artificial
intelligence is to reproduce human intelligence, function autonomously, and replace human
intelligence with computer systems.
Artificial intelligence has been successful to simulate autonomous applications such as
the IBM Deep Blue computer, which beat a chess grandmaster in 1997 (IBM 100, 2011). In 2011
Deep Blue was supplemented in some respect by the augmented intelligence of IBM’s Watson
computer, but it is not possible at this juncture to fully replicate the entire human intelligence.
There are limitations for simulating human intelligence. Although strength of artificial
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
4
intelligence is as a mathematically modeling machine for processing information, human
intelligence is superior for using intuition, interpreting context, and tolerating ambiguity in the
processing of information. When artificial intelligence conceptually reaches such human abilities,
an augmented intelligence approach would be useful for developing applications such as distance
learning.
Artificial intelligence, with its potentially disruptive feature of replacing human beings,
has overshadowed augmented intelligence and created trepidation among some experts including
theoretical physicist Stephen Hawking, Microsoft Founder Bill Gates, and others (Gates &
Hawking, 2015). Distance learning educators in particular have developed resistance for
adopting artificial intelligence applications that are perceived to take away the educators’ jobs
and act autonomously against educators’ plans. Shifting educators’ perspectives from artificial
intelligence to augmented intelligence may reduce resistance and enhance adoption of
applications involving augmented intelligence because augmented intelligence is grounded in
human-to-computer interaction. The goal is harnessing computers’ best strengths while keeping
human agents at the forefront, not replacing humans with computers. Such a view has been
shared by a number of organizations and resulted in developing applications with an augmented
intelligence approach.
Background
Distance learning is becoming increasingly prevalent in both higher education and K-12.
But is the technology of 2017 keeping pace with the increasing demands and tools for this mode
of instruction? Knowledge based artificial augmentation/intelligence may provide a partial
solution to this question. With knowledge-based systems (KBS), through the assistance of
subject matter experts, instructional designers, and programmers, it is possible to develop
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
5
applications for education (including distance learning) to support virtually any curriculum area;
as well as numerous administrative tasks and management tools. According to Bates, LaBrecque
& Fortner (2016), “Typically, students in online [distance learning] classes miss a higher
percentage of course material like assignments and assessments because of lack of consistent
communication and various computer and technology issues, often outside the control of the
instructor” (p. 6). Operating with a KBS may be one method to minimize the disparity by
providing virtually instantaneous feedback for most online curriculum areas, and harnessing the
ability of a KBS to be programmed with assignment instructions and even real time tutoring and
assistance.
Eras of Computing
According to Kelly (2015) three eras in computing developed: the tabulating systems era
(1900s-1940); programmable systems era (1950s-present); and cognitive computing era (2011????). Cognitive computing era made its debut in 2011when IBM’s Watson augmented
intelligence/KBS beat the greatest human champions of the game show Jeopardy. Since 2011,
IBM developers have been working actively in the field of augmented/KBS intelligence.
Augmented intelligence/KBS may assist humans to better analyze, visualize, and comprehend
the big data (Olshannikova, Ometov, Koucheryavy, & Olsson, 2015). In addition, Google
Knowledge Vault, Wolfram Alpha, Microsoft Minecraft (AIX), Facebook (FAIR Team), and
others are working to bring KBS to education. It may be time to take distance learning to a new
level by incorporating what KBS has to offer in the way of administrative, tutoring, feedback,
and research support.
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
6
Knowledge Based Systems and Big Data
A KBS is cognitive computing that understands natural language processing, adapts,
learns, and generates and evaluates hypotheses. The system can help in the area of distance
learning to assist instructors and students to make better decisions by penetrating the complexity
of big data. The capability cannot be understated. According to IBM system engineers, 90% of
the data in the world has been created in the last two years! (Rometty, 2016) and equates to 1.7
MB of data created for every person on the planet. In 2000 it was estimated that 500,000 people
participated in online purchasing. In 2016 that number rose to an estimated 20-25 million
purchases. According to projections by International Data Corporation (IDC, 2016) the digital
world will reach 40 zettabytes (ZB) by 2020. To put this number in perspective, consider that 40
ZB is equal to 57 times the amount of all the grains of sand on all the beaches on earth. The
quantity of data makes distance learning virtually unmanageable for keeping up with the growing
volume and velocity of information available.
Dark Data
Unstructured or unorganized data, termed dark data, is a significant issue in cyber-space.
According to IBM CEO Ginni Rometty (2016), 80% of the data in 2016 was dark and by 2020
this number will increase to 90%! From a librarian’s perspective Heidorn (2008) stated, “ . . .
presently dark data [are] not carefully indexed and stored so it becomes nearly invisible to
scientists and other potential users [distance learning students and faculty] and therefore is more
likely to remain underutilized and eventually lost” (p. 280).
Managing Big and Dark Data With Knowledge Based Systems In the Cloud
A KBS is one way to attempt the management of big and dark data by creating tools that
generate meaningful data. For example, using the cloud-based IBM Watson (Watson) platform,
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
7
all the known articles, texts, papers, presentations, and research studies on cancer have been
stored in Watson (Kohn et al., 2014). Physicians and other health care professionals can have
access to these materials in less than two seconds to acquire a potential diagnosis and
recommendations for treatment. Taking health care as the model, it is entirely possible to port
this technology to other areas including distance learning and education in general.
[PLACE FIGURE 1 ABOUT HERE]
Using KBS/Watson, tools can be developed to assist teachers, faculty, and students in the
areas of discovery, deep learning, large-scale math, fact checking, etc. In addition, time
management is a common challenge for many distance learners. KBS tools can provide real time
feedback to teachers, faculty, and students to help mitigate time management issues, along with
other problems and complexities surrounding distance learning.
Cognitive Computing and Knowledge Based Systems
Cognitive computing systems, including augmented intelligence/KBS, primarily work by
interacting and learning in a natural way with people to extend what both humans and computing
can do on their own. KBS is not thought of as artificial intelligence per se, although it has many
of the attributes; rather a KBS operates through augmented intelligence to help human experts
make better decisions by navigating the issues surrounding big and dark data (Kelly, 2015). In
the world of distance education managing big data can be accomplished by enhancing the
cognitive process of distance learning professionals, faculty, and students to help improve one’s
decision-making process in the moment. According to IBM’s Institute for Business Value
(2015), there is already a belief among subject-matter experts such as instructional designers that
current computer architectures and programming paradigms must advance cognitive computing
to include natural language processing.
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
8
Natural Language Processing
Assimilating the vast amount of information academics and others publish is a daunting
task for students and faculty, especially in the distance learning environment. Using sources such
as online libraries, Google Scholar, and other web-based sources can be challenging. Natural
Language Processing (NLP) uses natural language to mitigate such accessibility issues, but the
form of NLP developed for Watson, for example, is limited to English at this point. According
to Ferrucci, Levas, Sugato, Gondek, and Mueller (2012), NLP/Watson does not plot the inquiry
to a database of questions and simply search for the answer. As a form of machine learning
architecture, NLP/Watson performs by analyzing natural language content in both questions and
knowledge sources. Further, NLP/Watson (a) assesses probable responses, (b) tallies evidence
for those responses, (c) discovers and evaluates potential answers and (d) gathers and scores
evidence for those answers in amorphous sources. Sources include natural language (NL)
documents and systematized sources such as relational databases and knowledge bases. A
definition of NL is, “A human written or spoken language as opposed to a computer language”
(American Heritage, 2011).
Machine Learning and Application of Augmented Intelligence
Knowledge based systems use a hybrid form of machine learning. Machine learning can
be considered an extension of artificial intelligence that allows distance learning application
developers to automate analytical model building. This is accomplished by assembling
algorithms that can learn from and make estimates based on data (Intel, 2016). One of the
applications of augmented intelligence is the intelligent tutor system that refers to any computer
system containing interactive applications with some intelligence that facilitates the teaching and
learning process (Rodrigues, João, & Vaidya, 2010). Intelligent tutor systems encompass the
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
9
major components of one-to-one interaction between the instructor and learners, customized
instructions based on learners’ needs, and individual feedback for the learners. The systems have
been developed within various organizations during the last half a century. A recent metaanalysis conducted by Kulik and Fletcher (2016) from 50 controlled evaluations of intelligent
tutor systems indicated that use of the systems raised students’ performance well beyond the
level achieved by students taught with conventional methods by human tutors or another form of
computer tutoring. One of the recent successful intelligent tutor systems is Intelligent Tutoring
for Lifelong Learning (I-TUTOR) that is a multi-agent-based intelligent system to support online
teachers, trainers, instructional designers, and tutors. It is supported by the European
Commission and can be used in open source learning environments. In collaboration with
Pearson Learning, IBM’s Watson will be used as an intelligent tutor in online as well as
traditional classrooms (IBM Pearson, 2016).
Distance Learning and Knowledge Based Systems
Data from the National Center for Education Statistics (NCES) indicated that in the fall
of 2013 there were 5,522,194 students enrolled in distance-education courses at degree-granting
postsecondary institutions (NCES 2014). By comparison, K-12 high school distance education
enrollment was over 1.3 million from the period of 2009-2010 and constitutes an increase of over
1 million from 2004-2005, when enrollment was approximately 300,000. The 1.3 million
distance learners’ represents 53% of all the high school students in the U.S. (NCES, 2012).
According to a report published by the Babson Survey Research Group (2015), in conjunction
with Online Learning Consortium, Pearson, WECT, StudyPortals, and Tyton Partners, found
more than one in four college students take at least one distance learning course (total of
5,828,826) students, representing a year-to year increase of 217,275).
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
10
With this substantial influx of distance learners comes the challenge of disseminating
quality curricula and associated support tools. According to He, Cernusca, & Abdous (2011), a
majority of distance learning programs is unable to meet the emerging demand and requirements
for both students and instructors. In addition, there are also challenges to cognitive computing. A
key challenge for the advancement of cognitive computing or augmented intelligence will be the
availability of skilled humans as subject matter experts to further the development of cognitive
computing. Advancing cognitive computing capabilities and implementing cognitive systems
require unique skill sets, such as those of machine learning experts and natural language
processing scientists. These skills are currently in high demand and limited supply (Kelly 2015).
Using a distance learning/blended learning platform is one way to train students to meet these
job demands.
Knowledge Based Systems and Watson for Education
Knowledge-based systems development is at the cusp of emerging in education and
distance learning. Building on Watson’s health care model, IBM system engineers are applying
its cloud-based services to education. In late 2016 IBM released the mobile product IBM Watson
Element for Education. Using Watson, the mobile application provides educators with insights
on their students from a consolidated 360-degree view from a range of data sources (IBM
Education, 2016). The application uses a variety of data sources, including demographic,
academic, attendance, assessment, social and other data. Another benefit is that teachers can be
alerted to important information about student progress to facilitate provision of immediate
guidance if necessary.
In 2016, IBM released IBM Enlight For Educators, which consists of two components:
analytics and a library. The first component, IBM Watson Education Analytics “. . . uses a single
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
11
data representation of each student to provide the educator with a comprehensive profile” (IBM
Education, 2016). One creates the profile by gathering data from numerous sources, including
demographic characteristics, academic history, student interests, behavior, academic
strengths/weaknesses, etc. The second component is the IBM Watson Education Library.
Developers consolidated learning content to provide educators with a “simpler, rich source” of
materials (IBM Education, 2016). With this component, cognitive computing or KBS can
provide analysis that aligns with a curriculum and other resources including academic standards
and customized instruction for the individual student (IBM Education, 2016).
Both IBM Watson Element for Education and IBM Watson Education Library can be a
potential benefit for those teaching distance learning courses. One issue for many distance
learning educators is time. Tools developed by IBM, and the others may help with the demands
of time management. A second benefit using IBM Watson/KBS is managing and organization of
courses. Distance learning involves synchronous and asynchronous virtual instruction that
creates its own set of challenges when delivering, managing, and organizing distance learning.
Knowledge Based Systems and Microsoft®
Microsoft has entered the distance learning education arena with the online game
Minecraft™. Popular with ages ranging from elementary school to adults, Minecraft™ is an
example of where augmented intelligence and KBS are emerging. Microsoft Corporation® saw
the benefit for future cognitive programs by purchasing Minecraft™ in 2014 for $2.5 billion
dollars. In November 2016 Microsoft launched Minecraft: Education Edition. In the Minecraft:
Education website introducing the product, the writers point out what the platform has to offer,
including lesson plans, starter worlds (AI/Virtual) a place for educators to collaborate and a
mentoring program (Minecraft, (2016). The Minecraft™ mentor program connects educators
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
12
with others experienced in teaching with Minecraft™ to allow educators to explore ideas and
lesson plans created by educators covering topics from storytelling and poetry to city planning,
sustainable living and geometry.
Technology Behind Using IBM’s Watson for Development of Tools to Support Distance
Learning
Although cognitive learning platforms by various developers (e.g. Microsoft, Google,
Facebook, etc.) are emerging, the Watson expert system appears to be one of the best solutions to
facilitate the creation of a myriad of curriculum and administrative tools for distance learning as
well as ground-based applications. Is Watson an expert system? Some experts say that it is a
form of an expert system but goes beyond. Watson architecture is complex. This complexity
makes it challenging to determine in which category Watson fits, beyond IBM’s coined term of
augmented intelligence. Murdock (2013) believed that in some ways it could be classified as an
expert system.
Expert System
An expert system has a two-part structure consisting of an inference engine and a
knowledge base. The inference engine operates analogously to human reasoning by using the
facts presented to it by the knowledge base to infer the logical results of a proposition initiated
by the user. The inference engine acts on the data by applying “if-then” type rules. Every ifthen rule has multiple conditions and multiple actions from which to choose. Particular
conditions must be true for Watson to initiate specific actions. One deduces relationships
between the data presented and the actions Watson initiates by interpreting the first-order logicbased rules against the knowledge in the knowledge base (Ferrucci, Brown et al. 2010).
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
13
A knowledge base is a database that is structured into an ontologically consistent format
and is analogous to an encyclopedia. Ontological consistency of the knowledge base allows
Watson’s inference engine to make deductions that are valid and consistent within the particular
domain of instruction that is required. The knowledge base contains two types of knowledge:
expert and general. When a user queries an expert system like Watson, the inference engine
applies rules to the information contained in the knowledge base and new information is created.
After the Watson system verifies and validates the new knowledge, it is stored in the knowledge
base with the same ontological constancy with which the original knowledge was stored. This is
how expert systems like Watson learn and continue to grow to be increasingly more useful over
time.
Constructing a Knowledge Based Systems for Distance Learning
Constructing a KBS for distance learning using Watson may be a matter of consequence.
It may be that all the resources required for the construction of a KBS for distance learning are
not already an inherent part of the systems that make up Watson. If that is the case, then there
might be a need to build additional cognitive computing components with logical symbolic
relationships expressed as a series of rules. That type of programming requires the use of a
specialized language that is fashioned to use non-numeric or symbolic computing techniques
known as Prolog.
The name Prolog is derived from the words Programming Logic. Prolog is based on
formal logic and is considered to be a high-level programming language because it is machine
independent and can be read by humans. Prolog was built to search a knowledge base,
performing read and writes operations while preserving ontologically consistency. Prolog can list
facts and rules, define and solve logical expressions and is also considered to be a declarative
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
14
language. Prolog was used by IBM’s computer scientists, in part, to build the Watson system so
it is among the best, and possibly the only, computer language choice to use in extending
Watson’s cognitive computing capabilities (Michalik, 2011).
Extending the cognitive computing capabilities of Watson is a much different type of
programming than building new applications using Watson’s Application Programming Interface
(API). When you use Watson’s API you are building an application that draws upon established
application libraries that have known inputs, outputs and behaviors using a procedural language.
A procedural language is one that requires the programmer to construct a series of well-formed
and precisely described increments or steps. It requires a defined and well-controlled vector or
flow through the program. Our modern society is based upon this type of programming because
it is useful in performing tasks repeatedly given certain inputs; the computerized optimization of
an internal combustion engine for maximum performance is a relevant example of this concept.
When extending the cognitive computing capabilities of Watson using prolog you do not write it
as you would a procedural language, line by line. Prolog is a declarative language that has to be
written by describing a situation. The prolog compiler will tell you if the prolog sentence that
was written is true or not and it will also tell you what the variables are in that sentence without
the programmer having to declare them. This is a difficult style for many programmers to use let
alone master, but it must be utilized because it is the language that provides Watson with its most
valued capabilities.
The difficulty in using prolog mostly resides in how it is used to represent knowledge
about the world. Representing knowledge about subjects being taught via distance learning
requires an ontological understanding of the subject being taught and the data structures that
prolog requires in presenting that knowledge. With any given subject, the programmer must
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
15
understand the properties, types and interplay of the entities that exist for the domain of study.
This understanding when instantiated in prolog code will help prolog construct the variables
needed for the symbolic computations that need to be carried out for Watson to master the
domain of study or instruction.
Creating a useful ontology and successfully extending the cognitive computing
capabilities of Watson is not divorced from procedural programming. Functional programming
takes procedural programs and contains them in functions, as the name implies. Functions can be
used to build data structures that are used to control the flow, input and output of data.
Procedures contained in functions are required to ensure that ontologies are performed in
predictable ways that can be used to create distance learning applications.
When programmers first started programming with knowledge bases in ontologically
consistent formats, it became apparent that scaling would be difficult, if not impossible, unless
formats were modular with the ability to communicate. Programmers used ideas derived from
procedural and functional programming to encapsulate small and highly specialized knowledge
bases into small domain-specific units with the ability to communicate between the units. Such
functionality enabled programmers to use ontologically consistent knowledge bases as building
blocks of object-oriented code, thus providing unprecedented scalability.
[PLACE FIGURE 2 ABOUT HERE]
According to Füssl, Streitferdt, & Triebel (2015), Watson and KBS makes use of such
techniques for scalability that programmers can extend through “DeepQA” (p. 94). Deep QA has
an immense corresponding, component-based pipeline architecture that uses an extended set of
structured and unstructured content sources. The architecture supports a range of “pluggable
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
16
search and scoring components” for incorporation of several analytic techniques to apply to
distance learning (p. 95).
Use of Knowledge Based Systems in Distance Learning
Feasibility and development of KBS for use in distance learning is directly affected by
the tools available for use in extending the cognitive computing capabilities of Watson. To a
large extent the current tools are primitive for bringing together all the required programming
paradigms in developing a KBS for distance learning. Despite the limitations, Watson has come
a long way since winning the Jeopardy contest in 2011 with a cluster of 90 stand-alone servers.
In 2017 it is possible for a development team, including instructional designers, to purchase time
on the IBM Watson Developer Cloud without the added expense of owning servers (IBM
Developer, 2017). Creating the necessary tools required for the development of KBS
applications for distance learning is the first step. Educating programmers, instructional
designers, and subject matter experts to use the tools needed to develop a KBS for distance
learning is a necessary requirement. With augmented intelligence application development, the
human factor is present and needed.
Tools needed to develop a KBS for distance learning will not only include traditional
code-based tools. It is possible that specialized hardware also could be created. It is conceivable
that KBS applications for distance learning could drive specialized hardware that will allow
more rapid communication between devices on the Internet of Things (Cadavid, Ibarra, &
Salcedo, 2014). Specialized hardware for KBS applications in distance learning would be
required to speed up communications between the ontologies contained in the functions that
IBM’s Watson uses in its cognitive computing paradigm.
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
17
Precedence in Web Programming
Creating tools for the development of a KBS for distance learning and teaching
instructional designers and programmers to use the tools has precedence in web programming.
Web programming is the term associated with creating internet-based websites and internetbased applications and web programming has changed the way we live and work. Creation of the
first Internet web browser sparked a revolution in information-based programming changing the
way we shop, diagnoses illness and communicate with friends and family. Whole industries have
been created because of the capabilities of the web programming paradigm. According to Hanus
& Koschnickek (2014) web programming developed relatively quickly, within twenty years, to
become a driver of industry, health care, recreation and social interaction.
Knowledge Based Systems Beyond Web Programming
Development of the tools required to construct a KBS for distance learning may have an
equal or greater effect on society, as did the development of web programming. The
development of Watson may be one of the biggest events in computing history along with the
development of the Internet and Internet browser. The use of a KBS for distance learning
applications will not only enhance education in general but may allow individuals the
opportunity to learn and keep current in almost any type of profession with just an internet
connection. It will also enable individuals to obtain knowledge and understanding far more
quickly and at a younger age than has been possible in the past.
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
18
Feasibility Study: Knowledge Based Systems Using Watson for Distance Learning
Applications
Twenty instructional technology and computer science subject matter experts were
individually interviewed as part of a study to investigate whether developing KBS distance
learning applications was feasible. Questions presented to the participants were:
1. Can the knowledge-based system perform syntactic and semantic recognition in
support of the creation of a KBS for distance learning?
2. Do the unique challenges facing instructional designers and information technology
developers in producing knowledge-based programs/applications make a KBS
feasible to produce?
A second goal was to explore whether a conceptual framework exists in current academic
programs for incorporation into a development plan/model that can be migrated to a future KBS
scholarly writing solution. Given the current and emerging state of technology, 18 of 20
participants were confident that KBS could be migrated from health care to develop and support
curriculum and administrative distance learning applications.
With the information the participants provided, a prototype writing tutor application was
developed with the help of a group of subject matter experts consisting of an instructional
designer/distance learning faculty, computer programmer, and expert in the areas of grammar,
style, and mechanics. Researchers, working with IBM, are using the Watson Developer API
cloud-based system for this venture. It is hoped to have a beta version available in late 2017.
Conclusion
A continued focus on harnessing augmented intelligence/KBS to support educators in
distance learning is expected. For instance, IBM has plans to continue learning about human
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
19
intelligence and how people learn to be able to assist educators improve teaching and learning
processes using augmented intelligence applications. For IBM developers, a goal is to
“fundamentally bring the element of discovery, surprise and exploration back into the classroom,
and in the process, deeply engage the learner.” (Smith, 2014, p.1).
With the emerge of many augmented intelligence systems, instructional designers play
key roles in adopting augmented intelligence applications to identify the appropriate systems,
inform educators about the potential of augmented intelligence systems, and incorporate them
into the educational system and learning modules. Instructional designers should have multiple
roles as researcher, innovator, and informer. It is, perhaps, instructional designers’ responsibility
to shift the perspectives of educators from the philosophical concept of artificial intelligence to
augmented intelligence to reduce educators’ fear and hesitation in adopting augmented
intelligence applications. In addition, instructional designers should inform educators about the
limitations and strengths of augmented intelligence applications and their relation to human
intelligence. Knowledge based augmented intelligence can enhance the distance teaching and
learning process.
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
20
References
Ashby, R. (1956). Design for an intelligence-amplifier. In Automata Studies, C. E. Shannon and
J. McCarthy (Eds.), pp. 215-234. NJ: Princeton University Press.
Ashby, R. (1960). Design For a Brain. John Wiley & Sons, New York City, N. Y.
Bates, R., LaBrecque, B., Fortner, E. (2016). Teaching multiple online sections/courses: Tactics
and techniques [Electronic version] Online Journal of Distance Learning Administration,
19(3). Retrieved from
http://www.westga.edu/~distance/ojdla/fall193/bates_labrecque_fortner193.html
Babson Survey Research Group (2015). Online report card: Tracking online education in the
United States. Retrieved from
http://www.babson.edu/Academics/faculty/provost/Pages/babson-survey-researchgroup.aspx
Cadavid, A., Ibarra, G., & Salcedo, L. (2014). Using 3-D game technology in channel modeling.
IEEE Access (2) (pp. 1652-1669). DOI: 10.1109/ACCESS.2014.2370758
Engelbart, D.C. (1962). Augmenting Human Intellect: A Conceptual Framework. SRI Summary
Report AFOSR-3223; SRI Project No. 3578. Retrieved from
http://www.dougengelbart.org/pubs/augment-3906.html
Ferrucci, D., Levas, A., Sugato, B., Gondek, E., & Mueller, E. (2012). Watson: Beyond
Jeopardy! Artificial Intelligence, 99, 93-105. Retrieved from
http://dx.doi.org/10.1016/j.artint.2012.06.009.
Ferrucci, D., Brown, E., Chu-Carroll, J., Fan, J., Gondek, D., Kalyanpur, A. A., . . . Welty, C.
(2010). Building Watson: An overview of the DeepQA project. AI Magazine, 31(3), 5979. Retrieved from http://www.aaai.org/ojs/index.php/aimagazine/article/view/2303
Füssl, F. F., Streitferdt, D., & Triebel, A. (2015). Modeling knowledge bases for automated
decision making systems – A literature review. International Journal of Advanced Computer
Science and Applications, 6(9), 185-189. doi:10.14569/IJACSA.2015.060925
Gates, B. & Hawking, S. (2015). An open letter. Research priorities for robust and beneficial
artificial intelligence [open letter]. Future of Life Institute. Retrieved from:
http://futureoflife.org/ai-open-letter/
Hanus, M., & Koschnickek, S. (2014). An ER-based framework for declarative web
programming. Theory and Practice of Logic Programming, 14(3), 269-291.
doi:http://dx.doi.org/10.1017/S1471068412000385
He, W., Cernusca, D., & Abdous, M. (2011). Exploring cloud computing for distance learning.
[Electronic version]. Online Journal of Distance Learning Administration, 14(3).
Retrieved from http://www.westga.edu/~distance/ojdla/
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
21
Heidorn, P. B. (2008). Shedding light on the dark data in the long tail of science. Library
Trends, 57(2), 280-299. doi:10.1353/lib.0.0036. Retrieved from
http://muse.jhu.edu/issue/13721
IBM At 100 (2011). Deep Blue: Transforming the world cultural Impacts. Somers, NY. IBM
Corporation. Retrieved from http://www03.ibm.com/ibm/history/ibm100/us/en/icons/deepblue/
IBM Developer (2017). Build with Watson: Enable cognitive computing features in your app
using IBM Watson’s language, visions, speech, and data APIs. Somers, NY. IBM
Corporation Retrieved from https://www.ibm.com/watson/developercloud/
IBM Education (2016). Transform learning experiences with Watson: Personalized learning on
a whole new level. Somers, NY. IBM Corporation. Retrieved from
http://www.ibm.com/watson/education/
IBM Institute for Business Value (2015). Your cognitive future: How next-gen computing
changes the way we live and work. Somers, NY. IBM Corporation. Retrieved from
http://www-935.ibm.com/services/multimedia/GBE03642USEN.pdf
IBM Pearson (2016). IBM Watson Education and Pearson to drive cognitive learning
experiences for college students [Press release]. Somers, NY. IBM Corporation.
Retrieved from http://www-03.ibm.com/press/us/en/pressrelease/50842.wss
IDC (2014). The Digital Universe of Opportunities: Rich Data and the Increasing Value of the
Internet of Things [Executive Summary]. International Data Corporation. Retrieved from
https://www.emc.com/leadership/digital-universe/2014iview/executive-summary.htm
Intel (2016). Accelerate Intelligent Solutions with a Machine Learning Platform [Executive
Summary). Intel Corporation, Data Center High-Performance Computing. Retrieved
from http://www.intel.com/content/www/us/en/high-performance-computing/accelerateintelligent-solutions-with-machine-learning-platform-brief.html
Jabareen, Y. (2009). Building a conceptual framework: Philosophy, definitions, and procedure.
International Journal of Qualitative Methods, 84(4). Retrieved from
https://ejournals.library.ualberta.ca/index.php/IJQM/article/view/6118
Kelly III, J. (2015). IBM Computing, cognition and the future of knowing: How humans and
machines are forging a new age of understanding [White paper]. Retrieved from
https://www.research.ibm.com/software/IBMResearch/multimedia/Computing_Cognition
_WhitePaper.pdf
Kohn, M. S., Sun, J., Knoop, S., Shabo, A., Carmeli, B., Sow, D., … Rapp, W. (2014). IBM’s
Health Analytics and Clinical Decision Support. Yearbook of Medical Informatics, 9(1),
154–162. http://doi.org/10.15265/IY-2014-0002
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
22
Kulik, J. A. & Fletcher, D. (2016). Effectiveness of intelligent tutoring. Review of Educational
Research, 86(1), pp. 42 – 78. doi: 10.3102/0034654315581420
Lavenda, D. (2016). Artificial intelligence vs. intelligence augmentation. Network World
(Online), Retrieved from
http://search.proquest.com/docview/1809109037?accountid=458
Metz, R. (2010). Cloud computing explained. EDUCAUSE Quarterly, 33(2). Retrieved
from http://www.educause.edu/EDUCAUSE+Quarterly/EDUCAUSEQuarterlyMagazine
Volum/CloudComputingExplained/206526
Michalik, B. (2011, February 24). Watson and Prolog: or back to the future! Message posted to
https://www.ibm.com/developerworks/community/blogs/BernieMichalik/entry/watson_a
nd_prolog_or_back_to_the_future?lang
Merriam-Webster’s collegiate dictionary (10th ed.). (1999). Springfield, MA: Author.
Murdock, J. W. (2012) Is IBM's Watson an Expert System? Message posted to
https://www.quora.com/Is-IBMs-Watson-an-Expert-System
Minecraft: Education Edition Available Now! (2016). Retrieved from
https://education.minecraft.net/011116announce/
Natural Language. (2011) American Heritage® Dictionary of the English Language,(5th ed).
Retrieved from http://www.thefreedictionary.com/natural+language
Olshannikova, E, Ometov, A., Koucheryavy,Y. & Olsson, T. (2015). Visualizing Big Data with
augmented and virtual reality: challenges and research agenda. Journal of Big Data, 2(1),
2196-1115.
Rodrigues, J. J., João, P. F., & Vaidya, B. (2010). EduTutor: An Intelligent Tutor System for a
Learning Management System. International Journal of Distance Education
Technologies (IJDET), 8(4), 66-80. doi:10.4018/jdet.2010100105
Rometty, G. (2016, January 6). The Future of the Internet of Things is Cognitive. Address
presented at Consumer Electronics Show. Las Vegas, NV. Retrieved from
https://www.ibm.com/ibm/ginni/01_06_2016.html
Smith, F. (2014). EDUCAUSE 2014: What IBM’s Watson could bring to higher education.
EdTech. Retrieve from
http://www.edtechmagazine.com/higher/article/2014/10/educause-2014-what-ibm-swatson-could-bring-higher-education
United States Department of Education, National Center for Education Statistics. (2014). Digest
of Education Statistics, 2014 (NCES 2016-006),Table 311.15. Retrieved from
http://nces.ed.gov/pubs2014/2014067.pdf
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
23
United States Department of Education, National Center for Education Statistics. (2012). Fast
facts. National Center for Education
Statistics. https://nces.ed.gov/fastfacts/display.asp?id=79
United States Department of Education, National Center for Education Statistics. (2010). Fast
facts. National Center for Education Statistics. Retrieved
from http://nces.ed.gov/fastfacts/
Yan, L., Ma, X., & Wang, L, (2012). Intelligent interaction system of distance education on
natural language matching. In Jin, D. & Lin, S (Eds.), Advances in computer and control
systems (pp. 237-242). New York, NY: Springer.
Bios
Dale Crowe – is an associate faculty member at the University of Phoenix School of Advanced
Studies and a Research Fellow at the Center for Educational and Instructional Technology. His
PhD is in Instructional and Information Technology and also holds a Juris Doctor degree. Dr.
Crowe has been a technology consultant/IT project manager for K-12 schools, higher education,
corporations, health care, and museums for over 25 years. He was on the faculty, and a research
associate for two high research activity universities. He has presented at several domestic and
international conferences and is a peer reviewer of research textbooks for Sage Publications.
Martin La Pierre – is an Independent Computing infrastructure consultant at Ferentina, Inc.,
Doctoral Candidate, University of Phoenix, Col. USMC (Ret). He also holds a Master’s Degree
in Artificial Intelligence and Computer Science from De Paul University , a Masters of Business
Administration from National University and a Baccalaureate Degree in History from the
University of Illinois Chicago.
Mansureh Kebritchi is the founder and chair of the Center for Educational and Instructional
Technology Research, School of Advanced Studies, University of Phoenix, where she supervises
more than 200 faculty members to conduct research in the field of educational technology. She
holds a Ph.D. in Education, Instructional Technology and has a wealth of experiences in
managing faculty members to conduct research, teaching, mentoring doctoral students to
complete their dissertations, conducting research, and publishing and presenting the results in the
field of education. She is interested in studying innovative ways to improve quality of teaching
and learning in K-12, higher education, and corporate settings in online and face-to-face formats.
Some of Dr. Kebritchi’s recent publications address topics such as issues and challenges for
RUNNING HEAD:
KNOWLEDGE BASED ARTIFICAL AUGMENTATION/INTELLIGENCE
24
teaching successful online courses, facilitating styles of online instructors in course management
systems, pedagogical foundations of the educational computer games, factors affecting the
adoption of the computer games in school settings, and effects of mathematics games on
mathematics achievements and motivations of learners. She has been continuously publishing
and presenting her research results in international journals and conferences.