Conference PaperPDF Available

A study on the use of “Yams” for enterprise knowledge sharing

Authors:

Abstract and Figures

Text data understanding on social networking systems has become an important source of data for companies to understand their stakeholders better. The shift from pattern mining of structured database to non-structured text data has alerted companies to have a stronger presence in the new social media world. This research uses text miner module in the Statistical Analysis System (SAS) to analyze conversational data compiled from Yammer enterprise microblogging system. These inputs are used to study the topics discussed among employees. It is also used to examine the knowledge sharing activity among employees in the case company. This can be accomplished by analyzing the topics maps produced SAS software system. One is able to analyze the topic of discussion and the frequency of each topic on microblogging system platforms to observe the knowledge sharing and knowledge creation activity among employees. The case study company in this research project is a knowledge centric organization involves in knowledge sharing activity using a server-based Knowledge Management System (KMS). This research chooses employees that are involved in an active project. They will use Yammer instead of the current KMS system. The topic and text analysis diagrams are used to identify the patterns of discussion and the topics exchanged between employees. SAS (Statistical Analysis System) text mining tool is used to carry out the text mining analysis works where a number of visual representation graph were developed to study the communication patterns among employees. The results of this research had shown that text mining is able to surface employees' frequency of communication and topics of conversation through posting activities using Yammer in this research.
Content may be subject to copyright.
1
A Study on the Use of “Yams” for Enterprise
Knowledge Sharing
Angela Lee Siew Hoong , Tong Ming Lim, Soo Kar Leow and Justin Lim Renn Aun,
Sunway University
5, Jalan Universiti, Bandar Sunway,
Selangor, 46150 Malaysia
angelal@sunway.edu.my, tongmingl@sunway.edu.my, skleow@sunway.edu.my, juztin1002@hotmail.com,
Abstract – Text data understanding on social
networking systems has become an important source of
data for companies to understand their stakeholders
better. The shift from pattern mining of structured
database to non-structured text data has alerted
companies to have a stronger presence in the new social
media world. This research uses text miner module in
the Statistical Analysis System (SAS) to analyze
conversational data compiled from Yammer enterprise
microblogging system. These inputs are used to study
the topics discussed among employees. It is also used to
examine the knowledge sharing activity among
employees in the case company. This can be
accomplished by analyzing the topics maps produced
SAS software system. One is able to analyze the topic of
discussion and the frequency of each topic on
microblogging system platforms to observe the
knowledge sharing and knowledge creation activity
among employees. The case study company in this
research project is a knowledge centric organization
involves in knowledge sharing activity using a server-
based Knowledge Management System (KMS). This
research chooses employees that are involved in an
active project. They will use Yammer instead of the
current KMS system. The topic and text analysis
diagrams are used to identify the patterns of discussion
and the topics exchanged between employees. SAS
(Statistical Analysis System) text mining tool is used to
carry out the text mining analysis works where a
number of visual representation graph were developed
to study the communication patterns among employees.
The results of this research had shown that text mining
is able to surface employees’ frequency of
communication and topics of conversation through
posting activities using Yammer in this research.
Keywords – SAS, Topic and Text analysis, Text mining,
enterprise microblogging, usage pattern
I. Introduction
Facebook and Twitter are social media systems that
hold hidden knowledge which may be a source of
information for companies. It is crucial to understand
customers’ comments before investing into any
commercial activities. As a result, lot of research
works were conducted based on feedbacks and
comments mined from the unstructured data posted
by customers on microblogging platforms before any
important decision is made. However, mining text
information is not an easy task. One needs proper
tools to mine these data efficiently. In addition, the
amount of data produced by the bloggers and other
content producers are usually dynamic and this
makes text mining a challenging task indeed. In this
research, Yammer, an enterprise microblogging
system, is implemented for a shared services
company. The goal of this research is to study
knowledge creation and sharing activities in the
company. The factors of consideration on the choice
of company are based on the following criteria:
i. The company has started practicing
knowledge activity using some knowledge
management systems.
ii. The company sees value or needs where
knowledge make competitively different in a
globally open market.
iii. The company also sees organizational
knowledge as an asset for decision making,
improve productivity, performance and
quality from the perspectives of company
direction by the management.
As a conclusion, the case study company in this
research has fulfilled the criteria described above and
the management of the company has expressed
interest to improve its current knowledge activity
where the company’s goals have not met. The case
study company is a IT shared service company that
provides IT services ranging from consulting, infra-
structure, software development and system
integration for both internal and external
consumption.
In the research, the frequency of discussion and
topics of conversational posts by the employees of
the case study company will be studied. For data that
contain responses such as “yes” or “no” are easy to
analyze; however, extracting knowledge from the text
is a challenging issue. This is due to unstructured
2
nature of the data and the text intelligence that the
users want is not easy to extract from it.
Comprehensive text analysis requires thorough
understanding of the domain of study. Therefore, the
use of SAS Enterprise Miner (E-Miner) as the text
mining software to analyze the text data is able to
convert the conversational text into useful statistical
and visual representation.
In this study, a IT shared Services Company
was invited to participate in our research. The
research selects the type of company base on the
prescribed criteria explained in the last paragraph.
The chosen company enables us to study Yammer on
the participants’ usage and communication patterns.
In our research, Yammer was tested for a period of
30 days by on active projects. Entries entered by the
participants are analyzed to identify their topics of
discussion and the frequency of communication
among employees. Analysis on the concept maps or
concept linkings allows us to draw more accurate
research findings that reveal the communication
patterns of the employees. The text mining tool can
also identify whether the enterprise microblogging
system is a useful tool for knowledge sharing among
employees in the company.
II. Literature Reviews
Several related research papers were reviewed in
detail to solicit and understand theories and problems
unresolved from previous works. These existing
works provide better understanding on the use of text
mining in businesses. And these works also revealed
successful models and lessons on text mining as
guidelines for our project. Social media has become
ubiquitousand this is important for social networking
and content sharing. However, the content that is
generated from these websites remains largely
untapped. Sitaram Asur et al (2010) demonstrate how
social media content can be used to predict real-
world outcomes. In particular, we use the chatter
from Twitter.com to forecast box-office revenues for
movies. They also show that a simple model built
from the rate at which tweets are created about
particular topics can outperform market-based
predictors. In their works, they have further
demonstrated how sentiments extracted from Twitter
can be further utilized to improve the forecasting
power of social media.
Shimazu Hideo et al (2007) discussed the
impact of Web 2.0 on knowledge management (KM)
and the future orientation of KM in their paper. They
recognized that Web 2.0 has made its debut to
organizations with user participation-type culture has
expanded and the “collective intelligence” approach
has attracted attention of company operators. The
authors discussed the issue of knowledge provision, a
traditional issue of KM, and introduce a KM model
in the context of the Web 2.0 age that can expand
collective intelligence in a positive spiral by closely
linking it to knowledge extraction from various
communication tools and job systems.
Pattern analysis is able to identify and extract
subjective information from the source texts obtained
from the microblogging platform implemented in the
companies. Microsoft is one of the companies that
use text mining to conduct their analysis for their
online reporting application, NetScan. According to
Hari Mailvaganam (2007), the NetScan software uses
a combination of reporting, Online Analytical
Processing (OLAP) and Data Mining application to
analyze the posts from Usenet. Microsoft uses
NetScan to analyze their Usenet posts, frequencies, e-
mail addresses of posters, trend analysis, value of the
messages posted and eventual creation of a better
search engine for the company. However, the Usenet
posts are very chaotic and messy. Some information
is irrelevant and inaccurate. Therefore, NetScan is
used to analyze these posts to filter and obtain
meaningful information and sort the Usenet posts in
order. NetScan uses the text mining features of the
software to achieve its objective. It allows NetScan to
classify and cluster the system posts and generate a
predictive model using decision tree technique.
Reason on doing this is to evaluate which posts that
are useful.
Daniel Zeng et al (2010) observed the
importance of social media intelligence in their
editors' introduction and they claimed that social
media analytics research works have been facilitating
conversations and interaction between online
communities and extracting useful patterns and
intelligence to serve entities that include, but are not
limited to, active contributors in ongoing dialogues.
In spite of the latest development in the area of social
media related research, there are many definitions for
social intelligence and social media intelligence
provided by researchers and research labs with a
different twist. Social media intelligence as defined
by Nexalogy Environics (2011), on the other hand, is
“a series of analytical practices that, when combined,
allow for rich analysis of text-based data that goes far
beyond simple keyword counting, so-called
‘sentiment analysis’ and other such simple
measurements.” The definition provided by Hewlett-
Packard Development Company (2011) states that
“social intelligence seeks to understand, prioritize,
3
and leverage the data and behavioral insights that
social media provides”.
The definition of social computing
illustrated by Fei-Yue Wang et al (2007) is described
in Figure 1. Social computing represents a new
computing paradigm and an interdisciplinary research
and application field. Undoubtedly, it will strongly
influence system and software developments in the
years to come. Fei-Yue Wang et al (2007) expect that
social computing’s scope will continue to expand and
its applications to multiply. From both theoretical and
technological perspectives, social computing
technologies will move beyond social information
processing toward emphasizing social intelligence.
The move from social informatics to social
intelligence, as discussed by Fei-Yue Wang et al
(2007), can be achieved by modeling and analyzing
social behavior, by capturing human social dynamics,
and by creating artificial social agents and generating
and managing actionable social knowledge.
Figure 1. The theoretical underpinnings,
infrastructure, and applications of social
computing.
Ning Zhong et al (2010) stated that modern
computational technology approach such as Natural
Language Processing (NLP) technique was use to
mine the text documents so that human users are able
to understand the hidden pattern in the data. This
approach also allows the text mining engine to
discover relationships that exist between texts in the
files. In addition, Maria R. Lee et al (2007) pointed
out that, conventional knowledge management
through a centralized repository framework has been
the prominent approach to handle large volumes of
information since the instigation of World Wide Web.
However, the authors highlighted that knowledge
residing in the repositories has not been accumulated
or integrated to generate new intelligence. With the
massive increase of communication technologies
available, a change of paradigm in knowledge
management is produced. Their research investigates
the shift from a knowledge repository approach to a
conversational collaborative foundation of
knowledge management. Basic applications of
collaborative intelligence are proposed. It analyzes
recent web trends to produce support for the change.
III. Research Methodology
This research has adopted the following methodology.
To begin with, an active project that consists of
members from all the departments that participated in
the project was identified. This was followed by a
meeting to set the objectives of the test clearly. The
duration of the test was also identified and agreed
upon. The enterprise microblogging system chosen in
this exercise is Yammer. The microblogging system
was briefed, trained and implemented for a period of
30 days. Raw data of entries posted by participants
was stored in Excel files. The raw data is converted
into the SAS file format for the use of the text mining
system. The SAS Text Miner analyzes the
unstructured microblog text entries that consist of
employees’ opinions to identify and analyze topics of
discussion, linkings of topics, and communication
patterns among employees. Since the data contain
more than just simple “yes” or “no” answers or plain
numerical values, the need to process and analyze
such complex unstructured inputs is necessary in
order to provide human understandable outputs.
Among all, the text mining software is able
to identify similarities and differences from the
entries. The SAS software is able to mine and
produce frequency diagrams and topic linking maps
for the analysis purposes. The analysis allows
interpretation by researchers to identify the patterns
of the conversation among the employees of the
selected project as mentioned in early part of the
paper. The outputs generated by the text analytic
software provide a better and accurate understanding
of the communication patterns and usage pattern of
the employees of this exercise.
IV. Data Analysis
The analysis carried out in the research used a chain
of process nodes to construct a series of activity
nodes to analyze the raw text and generate analytic
outputs to be interpreted (Figure 2) by researchers.
These nodes are used to connect to the data source
which instructs the text miner to analyze the input
data to provide text and visual presentations.
4
Figure 1. Text Mining process nodes
Figure 3 depicts the communication and
interaction patterns of the post entries from the
employees who communicated using Yammer. Their
usage is presented as visual outputs. Their usage is
shown in the form of bar chart indicating frequency
of post by author. As illustrated in Figure 3, one of
the project members was the most active participant
among all the other participants in the microblogging
space. This member contributed the most Yams
entries in this exercise. The post contribution of the
most active knowledge creator has shown a
tremendous amount of support on the use of the
microblogging system. With this member’s
contribution as a knowledge creator and sharer, this
has also improved the assurance that the research
project will yield more anticipated outcome.
Figure 2. Frequency of post by participants
In Figure 4, on the other hand, showed that the term,
“license” had a value of 0.9 which was found to have
the highest weight. This simply means that the
importance of this term is the highest among all other
terms being used and the frequency of this term is not
very high, it has only five (5) occurrences in the
entries posted by the users.
Figure 4. Terms with Highest Weight
In addition to that, Figure 5 has clearly shown that
the frequency of terms and the weight of each term
highlight the communication concept and pattern and
the importance of topic discussed among employees.
For example, the box that is highlighted in red shows
“+server” term has the highest frequency count. It has
a total of 52 counts with the weight of 0.56547 that
appear in the data collected from the entire test period.
This clearly shows that the importance of this term is
at an average level. The higher the weight is the more
importance the term is from the entire corpus of the
text compiled in this experiment.
Figure 5. Terms with the Highest Frequency
In Figure 6, the concept maps are used to present the
discussion topics that are related to the term “ITSSC-
Information Technology Sunway Shared Center”.
Some of the terms could be expanded into another
level of detail by drilling into the next level of the
branch of that topic node to understand the topics –
sub topic of the communication. For example, the
terms “setup” in Figure 6, when it is expanded,
relevant topic is the infrastructure department. On the
other hands, the term “Putra” when it is expanded, it
shows that “Hotel”, “Sunway” and "Project” are the
related terms. Concept linking or topic map illustrates
topic of discussion, topic – sub topic relationships
and the frequency of such topic discussed among
employees through their weights. The thicknesses of
the lines that connect these terms is an indication of
the importance of the concept discussed among
5
employees. The lines connecting the term “Putra”
and “Hotel” is the thickest which means that the
frequency of these two terms is the highest. These
connections are relevant because our interview data
toward the end of the testing period has verified that
the finding of the patterns of the software has found
match with the interview data collected.
Figure 6. Concept Map for each term
In Figure 7, the concept map shows the topic
exchange and discussion pattern of the term “Putra”.
The “Putra” is a topic of discussion which originated
from the Putra Place project. It appears that “Putra” is
the second most relevant topic in the map. The topics
and their relationships show how one term relate
another term. The thickness of the lines in the
concept map (Figure 7) validates that the term “Putra”
is closely related to the term “hotel”. It also illustrates
the relationships of these terms. Furthermore, based
on the analysis of these entries, the concept map
shows that the second thickest term is the “place”.
This shows that the most related entries in the
knowledge network are originated from the terms
“join” and “knowledge” (Figure 7). This pattern
reflects the entries that were posted in Yammer from
the conversation of participants. Another example
that one observes in the diagram is the term “team”
where it relates to other terms such as “sales”,
“marketing”, “finance”, and “purchasing”. Based on
the conversational data in the Yammer, it was found
that there are few new groups that were created by
the users. It is clear that users actually make use of
Yammer to communicate among peers through newly
created groups for their own departments for ideas,
status and topic of discussion. Besides the analysis
that was conducted, another form of analysis was also
carried out by the text parsing and text filter nodes is
to extract information on specific terms used by the
participants. This is very useful to identify specific
information from a large amount of data such as data
on social media systems. Figure 7 shows the updated
activities sequence diagram.
Figure 7. Concept map for each specific term
Based on the activity diagram, new findings for the
information retrieval perspective are illustrated. By
using the Text Filter node, information can be easily
retrieved. Therefore, through the Interactive Filter
Viewer, the communication topics related to the
author are also discovered. When terms such as
“implementation” and “server” are entered as the
retrieving criteria, the documents on the screen shows
all the text entries which are contain such terms with
author detail will be presented visually.
V. Research Results Analysis
The most and least communicated topics were
illustrated through the use of concept map or concept
linkings diagrams. The use of topic map is to
illustrate the topics, occurrences and association of
topics discussed among employees whereas social
network graph illustrates actors or ties in a network,
relationships, connections and interactions of these
actors in the network. The patterns from the analysis
have helped the analyst to discover and identify the
communication topics among employees. From the
findings, the posting activities and the content of the
entries were analyzed. The analysis has also shown
that Phoebe Than had posted a fair amount of entries
which means that she has taken the knowledge
contribution champion role to create and share
knowledge with members of the project on Yammer.
Since the project is almost at its completion stage, the
most discussed topic was on the “legal” issues. The
patterns identified from the users’ postings on the
term “license” carries the highest weight. And the
term “+server” that is also having high frequency
matched with the content of the post-analysis
interviews. The reliability of the result is high
6
because licensing issue is the current topic in the
discussion. Furthermore, based on the observation
carried out on a day-to-day basis on the postings, the
term “+server” was always the most talked about
item among members. On the pattern analysis, the
topics of discussion were shown in the concept maps
which clearly display the knowledge exchange
pattern among employees. For instances, the concept
map on the term “legal privileges” is connected to
“unauthorized” and “legal”. This shows that these
topics discussed are related and relevant to the
project do take place. Furthermore, the text filtering
technique allows information retrieval through the
use of high frequency terms. By entering one or more
terms, the documents that have these terms will be
retrieved. It is clear that the text filtering can be used
to identify authors and their entries which are related
to the chosen terms. In a nutshell, the patterns of the
analysis show that the project members make use of
Yammer to share knowledge on the progress status of
the project. Although the testing period was short and
the number of post entries is small, the findings show
that entries posted are project related. Therefore,
microblogging would be useful to be used within
companies for knowledge sharing if it is given
sufficient resources such as time and larger group of
participants actively involved in this exercise.
VI. Conclusion and Future Research
In summary, the findings of the research have shown
that the communication patterns and topics of
discussion among employees based on the enterprise
microblogging system have shown signs of
improvement compare to the existing classical KMS.
The terms that have the highest occurrences and
weight are identified as the most important topics that
are frequently talked about among participants.
Furthermore, a number of concept maps were
developed to show the topics of discussion among
members of the project team. These topics were
selected to illustrate the significant increase of such
discussion and communication frequency between
employees. Lastly, the SAS tool can also be used to
retrieve specific text data based on key terms of
interest. This shows that text mining is able to
discover discussion patterns, terms and concepts
shared among employees. Throughout the research, it
had shown that the employees do use microblogging
to converse among each other and share information
with their co-workers. Therefore, microblogging is a
useful tool to share knowledge and information. The
sharing behavior and adoption of Web 2.0 tools on
employee productivity and job performance in
companies pave a strong foundation for future
research works.
References
1. Ning Zhong, Yuefeng Li & Sheng-Tang Wu,
(2010), ‘Effective Pattern Discovery for Text
Mining’ [online]
http://csdl.computer.org.ezproxy.sunway.edu.my
/dl/trans/tk/5555/01/ttk2010990136.pdf
[Accessed on 20th October 2011]
2. Hari Mailvaganam, (2007), ‘Evolution of
Analysis – Microsoft’s NetScan and Project Aura’
[online]
http://www.dwreview.com/Data_mining/Micros
oft_netscan.html [Accessed on 3rd June 2011]
3. StatSoft Electronic Statistics Textbook, (2011),
‘Text Mining’ [online]
http://www.statsoft.com/textbook/text-mining/
[Accessed on 6th June 2011]
4. Maria R. Lee and Yi-chen Lan (2007) "From
Web 2.0 to Conversational Knowledge
Management: Towards Collaborative
Intelligence", Journal of Entrepreneurship
Research June 2007, Vol.2, No.2, p. 47-62
5. Shimazu Hideo And Koike Shinichi (2007)
"KM2.0: Business Knowledge Sharing in the
Web 2.0 Age" NEC TECHNICAL JOURNAL
Vol.2 No.2/2007 Pp. 50-54
6. Kelsey Blair (2011) "How to Use Yammer… And
Why You Should Be Using It At Your Business"
[online] http://socialtimes.com/how-to-use-
yammer-and-why-you-should-be-using-it-at-
your-business_b40658 [Accessed on 10th March
2012]
7. Daniel Zeng, Hsinchun Chen, Robert Lusch and
Shu-Hsing Li (2010) "Social Media Analytics
and Intelligence", IEEE Computer Society Pp.
13-16
8. Sitaram Asur and Bernardo A. Huberman (2010)
"Predicting the Future with Social Media", IEEE
WICACM Vol. 1, Issue: 1, Pp. 492-499
9. Fei-Yue Wang, Daniel Zeng, Kathleen M. Carley
and Wenji Mao (2007) "Social Computing: From
Social Informatics to Social Intelligence", IEEE
Computer Society Pp. 79-83
10. Nexalogy Environics (2011) "Social Media
Intelligence: Taking Monitoring To Task", A
White Paper
11. Hewlett-Packard Development Company (2011),
"Unlock the Value of Social Media Data", A
White Paper
... Each customer has their own set of expectations and preferences, making it difficult for hotels to consistently fulfil those expectations. As a result, data analytics plays a vital role in this analysis with the hotel industry now turning to sophisticated analytical solutions for clues about how to please customers and keep them satisfied [9,10]. ...
... Many have developed and publicised the value of listening to customers' voices [10][11][12]. Many businesses rely on customer surveys or customer feedback forms to collect data in organised formats. ...
Article
Full-text available
With a large number of reviews being posted daily, online travel forums have become a very common platform for sharing travel information. Travel websites including TripAdvisor.com and Booking.com have become invaluable tools for hoteliers and travellers alike, enabling them to advertise hotel rooms, select hotels, and solicit and exchange reviews. This research paper uses data collected from TripAdvisor websites and 5501 reviews were collected from 4 different four-star hotels in Kuala Lumpur from 2012-2017. The work aims to extract the keyword (factors) that influence travellers in their choice of hotels using the framework of Visualizing Keyword Relationship Analysis (VKRA). The relationships of these keywords are illustrated using text analytics visualization techniques which are term frequency, keyword association and keyword network graph. The incorporation of these text analytics techniques offers insights for (i) travellers to view the opinions of other travellers on these hotels and (ii) hoteliers to understand travellers' preferences (location, room, services) and experiences, and improve their services in the future.
... With the spelling dictionary specified, the spellchecker will then list out the misspelt words from the highest to the lowest frequency. [This is just part of data cleansing, whereby the misspelt words may be manually corrected, or simply ignored] [21][22] [8]. Based on the co-occurrence of words in the documents, concept links are then formed, which helps in understanding the connections amongst words. ...
Article
Full-text available
Online travel forums have become an extremely popular platform for sharing travel information, with a large number of reviews being posted daily. Travel websites such as TripAdvisor and Booking.com have turned into very important resources for hotel operators and travellers alike, for promoting hotel rooms, choosing hotels as well as for soliciting and sharing feedback. Criticisms, compliments, dis-sensions, etc., are now accessible anytime and anywhere on the web, and can be readily amassed, while opinion mining techniques have developed rapidly. Together they provide the opportunity and capability to analyse and deduce factors that influence travellers in their choice of hotels. In this paper, we apply opinion mining on data collected from Tripadvisor websites. In total, 11,130 reviews on 4 hotels within the four-star and five-star categories in Kuala Lumpur are crawled, collected, and mined to identify the top-k most predominant information based on the most frequent and most related terms used in describing each of the chosen hotels. The results of this study would allow travellers to see the opinions of other travellers on these hotels, and hotel operators would be able to receive feedback to improve their services and in turn promote their hotels. This study is also carried out in view of future improvements in the techniques used and the analysis performed.
Chapter
The popularity of online platforms such as social media, review websites and blogs in recent years has increased the volume of user-generated content. Travel websites such as TripAdvisor, Agoda, Booking.com has been a popular companion to many travellers when deciding which hotels to stay and where to go in a foreign country. Travel websites such as TripAdvisor, Agoda and Booking.com are sources for consumers to obtain past experiences, reviews and recommendations from other travellers who have visited a specific hotel. Thus, organizations and businesses now turn to big data analytics to inspire them to discover hidden insights and gain better understanding. In this paper, we study several factors which are impacting customer satisfaction and to identify various hidden insights towards travellers’ experience through text mining.
Article
Effective knowledge exchange among software developers is crucial for the competitive performance of their organizations. Today, the constant pressure on businesses to continually innovate and the increasing capability of information technologies to facilitate broader and more distributed communication are driving organizations to leverage social media tools to improve performance. These tools, which have changed the way we share knowledge, enable people to connect, communicate, and collaborate. Research on knowledge sharing via social media is still in its early phases, with a comprehensive overview of the literature yet to be completed. Thus, using a systematic literature review approach, this study aims to map the current literature on the topic in relation to software development. Furthermore, this study highlights the findings of former research and identifies gaps in the literature. The study offers several insights for researchers and practitioners and proposes a future research agenda to strengthen knowledge in the field.
Article
Full-text available
Knowledge management represents a key issue for both information systems’ academics and practitioners, including those who have become disillusioned by actual results that fail to deliver on exaggerated promises and idealistic visions. Social software, a tremendous global success story, has prompted similarly high expectations regarding the ways in which organizations can improve their knowledge handling. But can these expectations be met, whether in academic research or the real world? This article seeks to identify current research trends and gaps, with a focus on social knowledge environments. The proposed research agenda features four focal challenges: semi-permeable organizations, social software in professional work settings, crowd knowledge, and cross-border knowledge management. Three solutions emerge as likely methods to address these challenges: design-oriented solutions, analytical solutions, and interdisciplinary dialogue.
Article
Full-text available
Conventional knowledge management through a centralized repository framework has been the prominent approach to handle large volumes of information since the instigation of World Wide Web. However, knowledge residing in the repositories has not been accumulated or integrated to generate new intelligence. With the massive increase of communication technologies available, a change of paradigm in knowledge management is produced. This article investigates the shift from a knowledge repository approach to a conversational collaborative foundation of knowledge management. Basic applications of collaborative intelligence are proposed. It analyzes recent web trends to produce support for the change. The article demonstrates the opportunity for more effective and feasible knowledge management.
Article
Full-text available
In a broad sense, social media refers to a conversational, distributed mode of content generation, dissemination, and communication among communities. Different from broadcast-based traditional and industrial media, social media has torn down the boundaries between authorship and readership, while the information consumption and dissemination process is becoming intrinsically intertwined with the process of generating and sharing information. This special issue samples the state of the art in social media analytics and intelligence research that has direct relevance to the AI subfield from either an methodological or domain perspective.
Article
Full-text available
In recent years, social media has become ubiquitous and important for social networking and content sharing. And yet, the content that is generated from these websites remains largely untapped. In this paper, we demonstrate how social media content can be used to predict real-world outcomes. In particular, we use the chatter from Twitter.com to forecast box-office revenues for movies. We show that a simple model built from the rate at which tweets are created about particular topics can outperform market-based predictors. We further demonstrate how sentiments extracted from Twitter can be further utilized to improve the forecasting power of social media.
Article
Full-text available
Social computing represents a new computing paradigm and an interdisciplinary research and application field. Undoubtedly, it strongly influences system and software developments in the years to come. We expect that social computing's scope continues to expand and its applications multiply. From both theoretical and technological perspectives, social computing technologies moves beyond social information processing towards emphasizing social intelligence. As we've discussed, the move from social informatics to social intelligence is achieved by modeling and analyzing social behavior, by capturing human social dynamics, and by creating artificial social agents and generating and managing actionable social knowledge
Article
This paper is intended to discuss the impact of Web 2.0 on knowledge management (KM) and the future orientation of KM. These days, the term KM is used rather less than hitherto. However, as Web 2.0 has made its debt and user participation-type culture has expanded, the new concept of "collective intelligence" has been attracting attention. Opinions are currently being advanced with regard to the concept and process of KM and the system architecture that can be used to implement it. This paper also deals with the issue of knowledge provision, a traditional issue of KM, and introduces a KM model in the context of the Web 2.0 age to can expand collective intelligence in a positive spiral by closely linking it to knowledge extraction from various communication tools and job systems.
Article
Many data mining techniques have been proposed for mining useful patterns in text documents. However, how to effectively use and update discovered patterns is still an open research issue, especially in the domain of text mining. Since most existing text mining methods adopted term-based approaches, they all suffer from the problems of polysemy and synonymy. Over the years, people have often held the hypothesis that pattern (or phrase)-based approaches should perform better than the term-based ones, but many experiments do not support this hypothesis. This paper presents an innovative and effective pattern discovery technique which includes the processes of pattern deploying and pattern evolving, to improve the effectiveness of using and updating discovered patterns for finding relevant and interesting information. Substantial experiments on RCV1 data collection and TREC topics demonstrate that the proposed solution achieves encouraging performance.
Evolution of Analysis-Microsoft's NetScan and Project Aura
  • Hari Mailvaganam
Hari Mailvaganam, (2007), 'Evolution of Analysis-Microsoft's NetScan and Project Aura' [online]
Unlock the Value of Social Media Data
  • Hewlett-Packard Development Company
Hewlett-Packard Development Company (2011), "Unlock the Value of Social Media Data", A White Paper
How to Use Yammer And Why You Should Be Using It At Your Business" [online] http://socialtimes.com/how-to-use- yammer-and-why-you-should-be-using-it-at-your-business-b40658 [Accessed on 10th
  • Kelsey Blair
How to Use Yammer… And Why You Should Be Using It At Your Business
  • Kelsey Blair
Kelsey Blair (2011) "How to Use Yammer… And Why You Should Be Using It At Your Business" [online] http://socialtimes.com/how-to-useyammer-and-why-you-should-be-using-it-atyour-business_b40658 [Accessed on 10th March 2012]