KNOWLEDGE BROWSER MONITORING SYSTEM (KBroMS)

Nur Farahin Musa1, Nur Syafira Hamzah2, Muhammad Awis Jamaluddin Johari3, Amru Yusrin
Amruddin4, Chew Yew Choong5

Software Development Lab, ICT Department, MIMOS Berhad, MALAYSIA

Abstract

Knowledge Management (KM) usually involves practices and strategies used in identifying, creating,
coordinating, combining, presenting, retrieving and distributing knowledge. Knowledge Management
System (KMS) refers to any kind of IT system that related to Knowledge Management. However,
traditional KMS is difficult to keep up with the rapidly growing information in the Internet. Thus, in this
study, the authors present a novel idea and also make use of some practical methods to capture, monitor
and measure the flow of knowledge in the Internet by using a new integrated knowledge management
platform. This platform, from here onward to be referred as Knowledge Browser and Monitoring System
(KBroMS). KBroMS is based on Nonakas SECI model that consists of four conversions which are
Socialization (tacit to tacit), Externalization (tacit to explicit), Combination (explicit to explicit) and
Internalization (explicit to tacit). In KBroMS, one or more components such as Forum, Blog, Chat, etc. are
mapped to one of the quadrant of SECI Model. By measuring the level of knowledge sharing between
each component, will able to capture and transform the activities into countable measurement which
represent the knowledge activity in each quadrant of SECI Model, hence giving user the ideas of how
knowledge transfer between tacit and explicit The author believed that system can be applied effectively
in knowledge management systems, which not only help users to visualize the sharing of knowledge but
most importantly indicate the effectiveness of KBroMS.

Keywords: SECI Model, Knowledge Management System, Knowledge Monitoring System, Knowledge
Browser

1. Introduction

Knowledge is something that can be information, description, facts or skills that someone acquired
through education or experience. Knowledge can be divided into two types which are tacit knowledge and
explicit knowledge (Smith, 2001). Tacit knowledge is the kind of knowledge that is difficult to written
into document or spoken like skills, ideas and experiences. Most of the tacit knowledge commonly could
be lack of incentives in place to document because of not easily to express it either using numeric or
words. Explicit knowledge is one kind of knowledge that is formal and systematic. It can be easily
communicated, shared and documented using numeric and words. It also can be a representation of
compounds that leaves the structure of its components implicit. Knowledge cannot be easily elaborated

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 178
discretely in term of number or an algorithm (Carver, 2001). Knowledge can flow from one entity to
another as example from tacit to explicit or from explicit to tacit (Ngah & Ibrahim, 2007). It is hard to
determine the flow as it is hard to elaborate and measure knowledge in the first place.

In this paper, the authors try to model knowledge using SECI Model proposed by Nonaka (Nonaka,
Toyama, & Konno, 2000). Through SECI Model, different measurement has been implemented to
measure knowledge in each respective SECI quadrant. Hence, the authors proposed a way to determine
the flow of knowledge between the SECI Models. SECI stands for Socialization, Externalization,
Combination, and Internalization. It was first proposed in 1995 by Nonaka & Takeuchi, though was
refined and expanded for a broader audience in the popular book The Knowledge Creating Company
(Nonaka, 2007).

Socialization is process that transfers tacit knowledge in one person to tacit knowledge in another person.
It is experiential, active and a living thing, involving capturing knowledge by walking around and
through direct interaction with customers and suppliers outside the organization and people.
Externalization is process for making tacit knowledge explicit. Dialogue is an important means for both.
During such face-to-face communication people share beliefs and learn how to better articulate their
thinking, though continuous feedback and the simultaneous exchange of ideas. Externalization is a process
among individuals within a group. Once knowledge is explicit, it can be transferred as explicit knowledge
through a process Nonaka calls Combination. This is the area where information technology is most
helpful, because explicit knowledge can be conveyed in documents, email, data bases, as well as through
meetings and briefings. Combination allows knowledge transfer among groups across organizations.
Internalization is the process of understanding and absorbing explicit knowledge in to tacit knowledge
held by the individual. Knowledge in the tacit form is actionable by the owner. Internalization is largely
experiential, in order to actualize concepts and methods, either through the actual doing or through
simulations. The internalization process transfers organization and group explicit knowledge to the
individual(Creation, Nonaka, Knowledge, & Knowledge, 1997).

Knowledge flow is important for each of organization to track information among member organization.
Knowledge flow is a complex process to realize and not easy to develop a tool to measure. This paper will
elaborate more on the idea to develop a system which from hence onward will referred to as Knowledge
Browser Monitoring System (KBroMS). KBroMS looks like other browser but is specialize more on
monitoring user activities in knowledge sharing and knowledge environment. By using KBroMS, user can
have a profile that will keep tracking all the activities that user performs in Internet and then will be shown
into a report which is based on pie chart and use SECI Model as reference to plotting. KBroMS is
developing on PHP and VB.Net programming language, while using MySQL as backend database server.

2. Methodology

This chapter at first introduces SECI Model and then explains the system and components of KBroMS.

2.1 SECI Model

SECI Model is widely used to representing knowledge in Knowledge Management domain. The four
knowledge conversion methods in SECI Model are Socialization, Externalization, Combination, and
Internalization. All these knowledge conversion methods arise due to continues interaction between tacit
and explicit knowledge (Cao, Yao, Li, Zhai & Zu, 2010). Besides, the effective collaboration of
socialization-externalization-combination-internalization in SECI Model has contributed to the success of

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 179
many business activities as it used adequate time for each conversion of knowledge either from tacit to
explicit, or explicit to tacit (Suzuki & Toyama, 2004).

Figure 1: SECI Model defined by Nonaka & Takeuchi, 1995

Figure 1 demonstrates SECI Model with dynamic knowledge conversion at each quadrant. The four
conversion methods involving tacit and explicit knowledge are as follows:
From tacit to tacit (i.e., Socialization)
From tacit to explicit (i.e., Externalization)
From explicit to explicit (i.e., Combination)
From explicit to tacit (i.e., Internalization)

2.2 KBroMS Platform

KBroMS is a knowledge browser which developed to determine the knowledge flow by monitoring the
user activities while browsing the Internet. This system consists of several components which are:
Blog
Chat
Forum
File Sharing
My Web Page
My Buddy List
Media Album
Report

2.2.1 Socialization

Socialization is described as the process to transform existing tacit knowledge into new tacit knowledge
(Chou & He, 2003). The knowledge sharing activities for instance sharing experience with individuals or
groups, social gathering, virtual learning, and face to face or informal meeting are considered as
socialization since it involves the process of communicating tacit knowledge with others.

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 180
In KBroMS, the Socialization process takes place when a user participates in chat. Besides, the interaction
between users through one to one or one to many chatting are also categorized as socialization. KBroMS
also allows file sharing between users. This activity is classified as Socialization as well.

2.2.2 Externalization

Externalization is the knowledge conversion method from tacit knowledge to explicit knowledge. In
simple way, Externalization is the process of gathering the knowledge that people obtains through
discussion or sharing session into documented explicit knowledge which accessible to all. Externalization
plays an important role to generate new knowledge from tacit knowledge (Nonaka & Takeuchi,1995).

Blog enable everyone to share and capture personal knowledge and extend the discussions across blog
(Chatti, Klamma, Jarke & Naeve, 2007). Apart from that, in context of social media, Externalization
process take place when user performed the activities such as blogging, commenting on blog post and etc.

Furthermore, new blog reader can contribute to the discussion by rewriting the blog post at their blog with
reference to the blog that they had contributed earlier, (Chatti, Klamma, Jarke & Naeve, 2007)
In KBroMS, the Externalization process take place when user creates new blog post about something that
they searched through the browser. Furthermore, when other user read the blog post and leaves comment,
the activity is categorized as Externalization.

2.2.3 Combination and Internalization

Combination is the knowledge conversion method from explicit knowledge to explicit knowledge.
Combination is defined as as the process of systematizing concepts into a knowledge system (Chatti,
Klamma, Jarke & Naeve, 2007). It combined different source of explicit knowledge which lead to the
formation of new knowledge.

In KBroMS, the Combination process can be seen in forum component when the comments that have been
posted contain new keywords repeatedly for several times and they have connection with the related topic.
Therefore, it can be considered as a new knowledge.

Internalization is the knowledge conversion method from explicit knowledge to tacit knowledge.
Internalization creates a practical knowledge through learning by doing and explicit knowledge would
help to accomplish the learning process (Cao, Yao, Li, Zhai & Zu, 2010). In context of computer
technology, Internalization process takes place when someone playing a computer games because they
tend to take risks and learn through trial and error method, (Chatti, Klamma, Jarke & Naeve, 2007).

Just like Combination, Internalization occurs in forum component of K-BRoMS. When new keyword
keeps on repeating for several times, it indicates the formation of new knowledge in the thread. However,
for Internalization, the user that involved in the thread will understand the new knowledge and
internalized it to form their belief.

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 181
 2.3 KBroMS System Flow

Figure 2: KBroMS system flow

Figure 2 displays the general flow of the system at each layer. At the presentation layer, user need to login
to the system before using the system. However, for new users they need to do the online registration. For
registration process, it requires user to validate the registration process through their email. After the
registration process completed, the user needs to login to the system and they can start browsing the
Internet as usual. The difference between typical Internet browser and KBroMS is that KBroMS comes
with features to capture and measure user knowledge activities on Internet based on SECI Model. Besides,
every contents of the URL will be analysed by Keyword and Content Extractor to measure the knowledge
flow and generate report for each user profile.

2.4 Measurements in KBroMS

Two types of measurement to be considered in KBroMS are quantitative and qualitative measurement.
Quantitative measurement is a measurement method that involves the collection of number. Whereas,
qualitative measurement is something that cannot be measured in number but concerned on the aspect of
quality. In KBroMS, the knowledge flow will be calculated based on the score given for Socialization,
Externalization, Combination, and Externalization categories. The quantitative and qualitative
measurements will be given as
Table 1:

Table 1: Measurements in KBroMS

Component List of Activities Quantitative Measurement Qualitative Measurement

1) Blog 2) Random user createo Number of blog post that user N/A
new blog post created
o Random user get 1 point

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 182
3) Blog 4) Reader read blog o Number of blog post that has N/A
post been read by reader
o Number of reader interested to
read the author's blog post
o Reader and author get 1 point

5) Blog 6) Author and reader o Length of comment given to o Step to determine the quality of
give comment to the blog post comments:
blog post o Length of comment that author1. Remove stop word and remove
reply to readers comment on duplicate word
his blog post 2. Identify whether the comment
o Reader and author get point related or not to the blog post
based on the quality of 3. If related, calculate the
comment remaining word and divide by
%10
ex :
remaining word = 11
11 % 10 = 1.
Therefore, point = 1
4. If not related, calculate the
remaining word and divide by
%20
7) Blog 8) Reader give rating o o The higher the number of star
Number of star icon that reader
for blog post click given by reader, the higher the
o Author get point based on quality of the blog post
number of star given by reader ;
1 star = 1 point, 2 stars = 2
points, and etc.
o
9) Chat One to one chattingo Number of word o Give point to both the user if
o Number of character conversation is relate with the
o Number of symbol room topic
o Number of stop word 1. Summary of conversation
2. Emoji
o
Chat One to many o Number of word o Give point to both the user if
chatting o Number of character conversation is relate with the
o Number of symbol room topic
o Number of stop word 1.Summary of conversation
2. Emoji
o
Forum Author and reader o Length of comment given to o Step to determine the quality of
give comment to the the topic of the forum comments:
topic at the o Length of comment that author1. Remove stop word and remove
comment's section reply to readers comment on duplicate word
in forum his forum topic 2. Identify whether the comment
o Reader and author will get related or not to the blog post
point based on the quality of 3. If related, calculate the
comment remaining word and divide by

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 183
 %10
ex :
remaining word = 11
11 % 10 = 1.
Therefore, point = 1
4. If not related, calculate the
remaining word and divide by
%20
Forum Reader give rating o o The higher the number of star
Number of star icon that reader
for overall forum click given by reader, the higher the
content o Author get point based on quality of the forum content
number of star given by reader ;
1 star = 1 point, 2 stars = 2
points, and etc.
Forum Read the commentso The author and reader will earn N/A
and feedback of the points for reading the
topic in the forum comments and feedbacks about
the topic

3. Experiment

3.1 Socialization

As mention in the previous chapter, socialization is knowledge conversion method from tacit knowledge
to tacit knowledge. Chatting is the activity that represented socialization in the quadrant of SECI Model.
The experiment for this activity is performed by making 3 users that are accessing to the same URL
discussing about the topic. Every conversation will be stored in database and evaluated by using
measurement as mention in methodology. The screenshot of the experiment is shown in Figure 3.

The measurement that has been decided is by using keyword, summary of conversation, and emotion that
the users used. If the keyword that users used in the conversation is same or have connection with
keyword in the topic of URL, users will get a point. Otherwise, the users will get nothing. In this
measurement, keyword extractor is used as a tool to find keyword in a sentence. To summarize the
conversation, whole conversation and point will be given for each participates. In this process, summarizer
tool is used to summary the conversation. Emotion is showing the feeling of users; it is subjective to
identify whether it is positive side or negative side. The point will be given for each of emotion that user
used. For this process, the author is using Emoji detector to detect Emoji that users used in the
conversation.

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 184
 Figure 3: Chat box activity

3.2 Externalization
Externalization is knowledge conversion method from tacit knowledge to explicit knowledge. In
KBroMS, the blog component is used to demonstrate the externalization process by carry out an
experiment involving User A and User B. Given a situation where User A inserts URL to KBroMS. After
the browser displays the URL content, User A read it and creates blog post about the content based on his
understanding. Meanwhile, User B searched the same URL like User A and he found out the blog post
that has been created by User A. Then, User B read the blog post and gives comment based on his opinion.

3.3 Combination
In this experiment, user A log on to KBroMS to search on Internet. User A input the URL in the search
bar of KBroMS. When user A surf the pages of returned search result, user A can open the new thread
based on the topic of the website. Say User B log on the system and visit the same website of user A. User
B can respond to the thread that user A has opened to be discussed on. User B must click on the topic in
order to give the feedback to the topic. If other users such a user C and D visit the same website, they can
respond to the user As thread as long as they click on the topic. They can post their feedback as many
they want to.

4. Result and Discussion

For Socialization in KBroMS, the result for chat box activity is shown as
Table 2. User amiradham got 2 points because one point for the summary of conversation and another one
point are for keyword that user use in conversation. User faridjohari got 2 points because one point for the
summary of conversation and another one point for the emotion that user used. User awisjohari also got 2
points because of summary of conversation and keyword that user used. User awisjohari does not get 3
points even though got two keywords in one conversation because the point is not show how much
keyword that are users are using, but did the user is using the keyword. The all results are depending on
the accuracy of the tools that are used for developing this system.

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 185
 Table 2: Result chat box activity

User Point
amiradham 2
faridjohari 2
awisjohari 2

Meanwhile, for Externalization measurement in KBROMS, point will be given to any user that creates
blog post. Therefore, user A will score 1 point because he created 1 blog post. Meanwhile, User B will get
point when they give comment to blog post. However, to determine the number of point that should be
given to User B, the comment will be evaluated based on quantitative and qualitative measurement. For
quantitative measurement, the length of comment will be taken into consideration. After counting the
number of word, the comment will undergo the qualitative measurement like mentioned in methodology
section.

Forum can be categorized as a Socialization and Combination part of SECI Model. When user A, B, C and
D post their message on the specific thread, each of them will get 1 point for each post. User A whom
open the thread will get another 1 point. All comments that have been posted will be detected whether it
contains new keywords repeatedly for several times and they have connection with the related topic, so it
can be considered as a new knowledge. The users that posted the new keyword will get another 1 point for
contributing the new knowledge which is Combination.

5. Conclusion

In this paper, SECI Model defined by Nonaka and Takeuchi has been used as a reference to measure the
knowledge flow in KBroMS. The user activities captured by KBroMS are mapped to each quadrant of
SECI Model. As a result, the flow of knowledge and the knowledge transfer between tacit and explicit can
be monitored and measured. By using KBroMS, it can leverage learning and knowledge sharing among
users. Besides, it encourages the user interaction through the blog, chat and forum component in the
browser.

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 186
References

Carver, J. (2001). Knowledge Flow, (November).

Proceeding of the International Conference on Artificial Intelligence in Computer Science and ICT
(AICS 2013) 25 -26 November 2013, Langkawi, MALAYSIA. (e-ISBN 978-967-11768-3-2).
Organized by WorldConferences.net 187