http://www.diva-portal.org
Postprint
This is the accepted version of a paper presented at The Scandinavian Research Seminar on Information and Decision Networks, 22-24 April 1994, Växjö University, Växjö, Sweden.
Citation for the original published paper:
Eberhagen, N. (1994)
Media choice and social networks in the information selection process.
In: Lundberg, B.G. (ed.), Proceeding of The Scandinavian Research Seminar on Information and Decision Networks Växjö, Sweden: Department of Mathematics, Computer Science and Statistics, Växjö University
N.B. When citing this work, cite the original published paper.
Permanent link to this version:
http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-6370
Media choice and social networks in the information selection process
Niclas Eberhagen
Department of Mathematics, Statistics and Computer Science Växjö University
Box 5053, S-351 95 VÄXJÖ, SWEDEN E-mail: Niclas.Eberhagen@MASDA.HV.SE
Abstract
This is a position paper presenting the background to my current studies of information filtering as a mean to cope with information overload and to ensure the quality of the decision situation, and my ambitions concerning future directions for studies, such as focusing on media choice and usage social networks in information filtering processes.
Keywords: information overload; information filtering; social filtering and social networks.
Introduction
This paper briefly presents the background to what my studies are, what I have achieved so far and what direction I will pursue in further studies.
I am undertaking post-graduate studies within a Ph.D. program, to which I became a member in the spring of 1992 at the Royal University of Stockholm, within the field of Information Systems’
Science. I plan to present my Ph.D. Thesis in the spring of 1998. The focus of my research studies is on information selection within an organizational context.
The background for these studies can be derived from the work by both Huber (1984) as well as Drucker (1988).
Huber (1984) characterize the post-industrial society as experiencing greater levels of knowledge, complexity, and turbulence, and that each of these will be increasing at a considerable rate. The amount of available information will grow and its absolute growth will increase. The increase of knowledge will lead to large increases in technological, economical, and social specialization and diversity. The high diversity and specialization will lead to large increase in societal
interdependences and thus escalate the level of complexity and its absolute growth. The increase of turbulence follows from the rapidity of events and increasing knowledge, causing many
technologies to be more effective and shorten the duration of the events, thus permitting more events per time unit. Effects of advanced communication technologies and computing technologies, will be that they increase in availability to the individuals, and that they will open up new sources of information which originally were external to the organization. Decision making will be more frequent, faster, and complex thereby increasing the decision-task loads greater than before, putting
high demands on the acquisition and distribution of the available information, and the guarding against information overload as the amount of available information will increase.
Drucker (1988) points out that the organizational structure is changing, becoming more flatter and decentralized, leading to an increase on the information load and the decision-task load of the management.
Feldman (1981) further shows that individuals within the organization tends to gather more
information than necessary for the decision-situation at hand and that much of this information has a weak link to the decision-situation thus adding on to the problem of information overload.
It is here the concept of information filtering enters the stage. Information filtering as an activity is defined as the process by which individuals scans different channels for distributing information by matching their needs with the information at hand. The channels as well as the sources can be of both external and internal origin.
In a study, made by Malone, Grant, Lai, Rao, and Rosenblitt (1987), of different types filtering processes, three general processes which individuals used in order to select information were identified:
- cognitive filtering process, which is mainly based on a characterization of the content of the distributed information by means of keywords and matching the individuals information needs, also characterized by keywords. Example of such a system is the Info Lens system (Malone et al. 1987).
- social filtering is based upon the relations that exists between individuals in an organization or group. The relation between the distributor and receiver often dictates how the receiver will classify the information. The closest example of such a system is the Tapestry system (Goldberg, Nichols, Oki, and Terry 1992) which is based upon a filtering process that matches the information needs of an individual with the annotations made by other individuals regarding the information.
- economic filtering; the receivers of information tries to value the effort it would to take to digest it. If the information is too rich in volume the receiver may ignore it due to the time it would take to process it.
Current work
It is these types of information filtering processes my studies has come to focus upon. My
achievements so far has been to study and explore different techniques and methods for information selection.
I started my post-graduate studies (Eberhagen 1992) by exploring techniques for information organization according to the principle of heuristic clustering (Salton 1989) to group information which was related to each other by content. The study here aimed to explore the characterization of information by using keywords from the content of the document. The sets of keywords of different documents were matched with each other and if a sufficient level of similarity existed between two documents they were considered to be related and therefore grouped together.
I also explored techniques for cognitive information filtering (Eberhagen 1993). The techniques I have explored mainly covered the representation and matching of information needs. These are
roughly the same techniques as used within Information Retrieval since they basically are “two sides of the same coin” (Belkin and Croft 1992).
Instead of matching two lists of keywords from two different documents, a list describing the content of the information needs, the selection criteria, was used. If two lists of keywords, when matched with each other, showed a sufficiently high level of similarity the document was selected and considered relevant to the information needs. Inspiration to this work was the Info Lens system (Malone et al. 1987).
Results from both studies showed that the “sufficiently high level of similarity” lay between 15% to 20% of overlap between the two lists. This might seem to be a low figure but in order to
characterize the content of a document one must use many keywords to cover all aspects. This means that there are many more keywords that need to matched and a higher chance of establishing a mismatch if the threshold is too high.
One must remember that this matching involves no actual understanding of the text but rather a form of pattern matching of words. This implies that the process of establishing a match gives us rough estimates which calls for a need of exploring techniques such as the principles of relevance feedback (Salton 1989) that can help in fine tuning the list of keywords describing the information needs.
Further study
However, I aim now to raise my focus, within my studies, from the rather technical aspect, concerning the representation and matching of the information needs, which I have had when studying the information filtering processes.
When looking upon the work of Malone et al (1987) and their system Info Lens one thing is strikingly clear; this kind of system is a text-based filtering system which only reacts on keywords for its matching purposes. The same applies for the Tapestry system (Goldberg 1992). Both are based on formal information channels for distributing text-based documents. Both types of filtering processes are based on a rather poor media for distributing information (Daft, Lengel and Trevino 1987).
Thus it seams that the choice of a text-based media, which effectively ignores other types of signals of information, save the textual represented, could be used as a means for filtering information.
Concentrating on a certain type of media and ignoring the richness of cues, which can come from a choice of a different media for the same type of information, can very well be of importance when cooping with problems of information overload.
It has been shown (Daft, Lengel and Trevino 1987) that choice of media for distributing
information can be crucial depending upon the situation and whether the purpose of information distribution or scanning is to reduce uncertainty or to resolve equivocality.
The choice of media when scanning for information could very well be a filtering criteria. Using rich media, with a high bandwidth, when scanning for information could as easily add on to the problem of information overload due to an increase of different signals and cues, but the presence of such can on the other side as easily help to characterize the information more quickly as well as allowing a higher degree feedback, depending on the nature of the media, in order to fine tune ones criteria for selection.
Another aspect of the filtering processes covers the social filtering process. Social filtering as defined by Malone, Grant and Turbak differs somewhat from how it is applied within the Tapestry system (Goldberg 1992). Social filtering as defined by Malone et al. (1987) covers the aspects of whom people turn to in order to find information, individuals within the organizations has some knowledge of what information their neighbors think is useful and what sources of information they utilize.
Social filtering is based upon the informal social network of the individual, covering both advice network, showing who he turns to for advice i.e. who does he trust, and friendship network, showing who he considers to be his friends and thus the ones he turns to in times of crisis (Krackhardt and Hanson 1993).
The social network can then be used both to show the structure of informal communication as well as the process of informal communication, by stating who participate in an information exchange process and how it is done. By mapping the network of social contacts it can be used as a model showing the individuals’ social filtering processes and structures.
Such a model uncovering the picture of the organizational social network, showing both
gatekeepers and competency centers (Krackhardt and Hanson 1993), also shows that many of the information sources will be informal or derived from informal structures and of human character.
Within a selection process they might be scanned by many different types of media, ranging from face-to-face to email depending on the situation in which the individuals finds himself (Daft, Lengel Trevino 1987).
When developing any system which is to aid individuals in their information selection process one must take into consideration that it is not only the goals, and thus the information needs, of the individuals that will vary between individuals, but also the structure of the information selection process as the network of informal communication may change due to varying information needs.
Thus care must be taken to ensure that usage of IS support not imposes to much structure (Alter 1992) to extent that it will not allow the dynamics of a social network function freely within the framework of the IS.
To summarize the different aspect of information filtering that I have discussed so far I present below two directions of further studies:
- Firstly; the study of the usage of media within information selection processes. Here it is possible to incorporate such theories as semiotics, information theory, and media richness theory in order to build a better theoretical basis for understanding the usage media and its effect upon the
information selection processes in different decision situation. There is also a need for establishing a database covering different aspect of information selection and media usage, within an
organizational context, in order to build a basis from which conclusions can be drawn and new questions may arise.
- Secondly; the study of informal social networks as means of uncovering the structure and process of the information selection process is here another direction of further research. Incorporating both social network theory, communication theory, and social information processing theory in order to uncover, as above mentioned, models of social filtering in order to establish possible IS support.
References
Alter, S. (1992) Why Persist with DSS when the Real Issue Is Improving Decision Making?
DECISION SUPPORT SYSTEMS: Experiences and Expectation, Elsevier Science Publishers B.V., North-Holland, 1992.
Belkin, N.J., and Croft, W.B. (1992) Information Filtering and Information Retrieval: Two Sides of the Same Coin?, Communication of the ACM, Vol. 35, No. 12, December.
Daft, R.L., Lengel, R.H., and Trevino, L.K. (1987) Message Equivocality, Media Selection, and Manager Performance: Implications for Informations Systems, MIS Quarterly, September.
Eberhagen, N. (1992) Automatisk Dokument Klassifikation: Heuristisk Klustring, Examensarbete 10p för Systemvetenskaplig linje, Högskolan i Växjö, Maj.
Eberhagen, N. (1993) Information Filtering, Working paper of the Information System Science’s group, ISV-WP-15.
Feldman, M.S, and J.G. (1981) Information in organizations as signal and symbol, Administrative Science Quarterly, Vol. 26, March.
Goldberg, D., Nichols, D., Oki, B.M., and Terry, D. (1992) Using Collaborative Filtering to Weave an Information Tapestry, Communication of the ACM, Vol. 35, No. 12, December.
Huber, G.P. (1984) The Nature and Design of Post-Industrial Organizations, Management Science, Vol. 30, No. 8, August.
Krackhardt, D., and Hanson, J.R. (1993) Informal Networks: The Company Behind the Chart, HARVARD BUSINESS REVIEW, July-August.
Malone, T.W., Grant R.K., Lai, K-Y., Rao, R., and Rosenblitt, D. (1987) Semistructured Messages Are Surprisingly Useful for Computer-Supported Coordination, ACM Transactions on Office Information Systems, Vol. 5, No. 2, April.
Salton, G. (1989) Automatic Text Processing; The Transformation, Analysis, and Retrieval of Information by Computer. Addison Wesley, USA.
