Theme 3 (pre)

The Journal of Information Science

I have chosen the Journal of Information Science that has an impact factor of 1,158. The journal describes itself as “a peer-reviewed international journal of high repute covering topics of interest to all those researching and working in information science and knowledge management”. The journal focuses on how to analyze and visualize information of different kinds. With articles such as “Keyword extraction for blogs based on content richness” and “Hierarchical graph maps for visualization of collaborative recommender systems” I find the journal to be very interesting. I think an important factor in media technology is the ability to handle large amounts of information and also be able to visualize that information in a way that not only people within the subject understand. Therefore I find this journal to be relevant, although not limited, to media technology.

Keyword extraction for blogs based on content richness

The paper I choose is called “Keyword extraction for blogs based on content richness” written by Jinhee Park, Jaekwang Kim and Jee-Hyong Lee. The papers aim is to examine a new method to extract topic keywords of blogs, based on the richness of content. They talk about how blogs today is an important way to distribute information and news on the Internet and about the problem that exists in finding relevant information among the thousands of blogs and blog posts. They mention different methods that already exists and why these aren’t ideal. Their solution is to extract keywords from blogs using a new measure, richness, which indicates how much a blog covers the “trendy subtopics of a keyword”. To obtain these trendy subtopics of the chosen keywords they used the web to find trendy and popular content related to a topic or keyword there. With the trendy information about a topic they could choose a keyword that better match what is popular right now, much better so than previous methods. This is because they measured the richness of blogs using the web context previously mentioned.

The point of this is to help readers understand the content of different blogs more easily and have an easier time finding the relevant information they are looking for. In the paper they compare the results with other methods for keyword extraction which are based on a statistical method and could see that the proposed method gave superior results in hit counts, trendiness and consistency.

I think this paper is very interesting and also very relevant for today’s society. We are a society based around Internet and the Internet is not getting smaller, rather it seems to grow exponentially. Blogs are an excellent example of what the Internet is doing with information distribution. Today everyone can be a “news reporter” and in this sea of information we need methods to navigate and find what we are looking for. Their method sounds great since they take the most important factor into account, the web itself. By looking at what’s trending right now on the web they make the information later presented relevant to the “here and now” which I think is really clever.

I might not agree with everything they do however, an example of this is the assumption they to that a topic in general consists of 1000 unique words. That seems like something that could vary a lot, especially when you consider the amounts of blogs, blog posts and topics there are on the web and that it is constantly growing. However I’m not an expert at these sorts of things so who knows, it might be an accurate assumption.

I would like to see if this is applicable on other things on the web rather than just blogs. I understand that blogs have a very easy to work with format but it would be interesting to see if you could generate “keywords” on other things online rather than just blogs.

1.     Briefly explain to a first year university student what theory is, and what theory is not.

Theory is not data, graphs, diagrams, lists of data or hypotheses but rather the how, why and when something occurs. Basically the explanation of why the data, graphs and so on are interesting and how different parts of the paper are connected. The data in itself doesn’t generate the theory; this is where the researchers come into the picture because it is they who generate the theory itself.

2.     Describe the major theory or theories that are used in your selected paper. Which theory type (see Table 2 in Gregor) can the theory or theories be characterized as?

I would say that the major theory in the paper I choose is the theory for design and action. Since the paper is about how this new way of generating keywords is used to evaluate and prove that their method is superior, therefore proving their hypothesis. All in all the theory for design and action is about how to do something and in this case, how to do something better.

3.     Which are the benefits and limitations of using the selected theory or theories?

The benefits with the theory for design and action is that it says how to do something because you can get an answer to the questions “what?”, “when?”, “how?”, and “where?”. In the text Gregor mentions that there are some limitations and that “the user requirements include a need to translate expert knowledge into actionable knowledge for non-experts” (table 8).