=Paper=
{{Paper
|id=Vol-443/paper-9
|storemode=property
|title=Rich Interfaces for Browsing News in Blog Posts
|pdfUrl=https://ceur-ws.org/Vol-443/paper9.pdf
|volume=Vol-443
}}
==Rich Interfaces for Browsing News in Blog Posts==
https://ceur-ws.org/Vol-443/paper9.pdf
Rich Interfaces for Browsing News in Blog Posts
Earl J. Wagner Jiahui Liu Lawrence Birnbaum
ewagner j-liu2 birnbaum
@u.northwestern.edu @northwestern.edu @cs.northwestern.edu
Northwestern University, Evanston IL USA
ABSTRACT Often background events are related to the event covered by
Semantic models of news can enable richer interfaces for a news article by being part of the same overall situation -
end-users to learn the context of news events referenced in perhaps an earlier event in the situation, where by situation
blog posts. We present Brussell, a system that uses content- we mean a limited sequence of causally-related events, such
specific models of news event situations to perform as all of the newsworthy actions in a lawsuit. For example,
anticipatory information retrieval, organize extraction an account of a lawsuit being settled by the defendant may
results and present a novel, structured interface for refer back to the original filing of the lawsuit and both
navigating among the events of a news situation. events are part of a particular lawsuit situation. Alternately,
a news article may reference other similar or related
INTRODUCTION situations. A similar lawsuit may be taking place
A blogger commenting upon a news event in a blog post is simultaneously in another locale. Related lawsuits include
likely to link to a relevant news article, often in the first a suit acting as a case precedent, or other suits involving
sentence of the post. By providing a link, the author can some of the same participants, such as other suits against
assume that the reader has read the linked article and is the defendant.
familiar with the news event it covers. In fact, often the
“value-add” for blog posts is not in repeating the All of these relationships are part of the situational context
information of the article, nor reporting new information, that the user draws upon in making sense of the news event
but in sharing the blogger’s opinion or analysis. Bloggers a news article describes. This context gives rise to specific
use the news article as the starting-point for further questions, such as:
discussion. • What happened in this situation?
But what happens when the blog reader is unfamiliar with • How did this situation start? How did it end?
the news event described in the article? Often, a blogger
will argue that a current event is caused by, or part of a • What happened in the other situations referenced in
trend involving, past events. Alternately, if the event that this article?
the blog post refers to is not recent, or if the post itself is • What other similar and related situations have these
old, the reader may be interested in finding out its outcome. participants been involved in?
We argue that software can help the user by explaining the
context of news events through content-specific structured Neither conventional software and web sites for blogs and
presentations. news, nor research systems for “semantic-blogging”
provide content-specific support for answering these
BACKGROUND TO NEWS EVENTS IN BLOG POSTS questions, however.
The issue of explaining the context of news highlights one Without an in-page link, the user must find related articles
important difference between blog posts and news articles. manually in order to answer questions like these. She must
The blog format enables authors to write in a conversational identify relevant terms such as entity names and situation
style and presume familiarity with earlier commentary. keywords. Then she cut-and-pastes them into a news
Posts generally don’t provide as much context as news search engine. Finally she sorts through lists of results to
articles and are often written with the expectation that find relevant articles. These steps make for an inconvenient
readers will have been following the blog and are familiar process familiar to anyone who reads news on the web.
with the events they discuss. Journalists writing news Even news timelines provided by advanced search engines
articles, on the other hand, provide “background” to current are unable to provide content-specific overviews of a
news events by explaining the events and factors that led to situation in accordance with the user’s expectations of how
their occurrence. it begins and continues.
Workshop on Visual Interfaces to the Social and the Semantic Web We present Brussell, a system that performs anticipatory
(VISSW2009), IUI2009, Feb 8 2009, Sanibel Island, Florida, USA. information retrieval and model-based information
Copyright is held by the author/owner(s). extraction to support the user in exploring the situational
1
context of the news. Brussell retrieves news articles and
extracts information to create models of news situations.
When a user selects a situation, it presents a storyline with
the major milestone events. Clicking on the event label
loads an article that either immediately covers the event or
is the earliest mention of the event. Evidence that an event
took place, for its date and location, or for important
attributes of participating entities can also be viewed in the
form of collected textual snippets and links to source pages.
All of these features work together to provide the user a
“big picture” view of the news situation and answer high-
level questions about it.
EXAMPLE
Consider the case of a user reading about the kidnapping of
a BBC journalist on a blog post. Although the user is
vaguely aware of this incident, he would like to find out
more. With standard search technology, he would enter Figure 1. Viewing a reference to a news situation
terms into a search engine and peruse the results in order to within a blog post.
develop an overall sense of how the kidnapping situation
transpired. Through Brussell, he can interact directly with
the text referencing the situation. Moving the mouse over
the text causes it to be highlighted (see Figure 1).
Right-clicking on this highlighted text opens a context
menu presenting options for viewing the history of the
situation and finding out more about its participants (see
Figure 2). The user wants to see a summary of what
happened, so he selects the first option, which updates the
toolbar to show a storyline for the kidnapping with its major Figure 2. Asking about the situation.
events and their dates (see Figure 3). With this high-level
view of the kidnapping situation, he is able to see the how
the situation unfolded through the occurrence of its
milestone events. Alternately, selecting one of the menu
items to learn more about a participant would load a similar
timeline view presenting all of the situations the participant
has been involved in.
Next, he wants to know exactly how the situation ended, so
he selects the "release" event button that loads the most
Figure 3. Viewing a storyline for the selected situation
relevant page describing the event in detail (see Figure 4).
By reading this page the user can learn more about the
circumstances under which the journalist was released.
ARCHITECTURE
Brussell consists of a Firefox browser plugin and server
software, which may both run on the same computer.
When the user visits a web page the browser plugin sends
the current page title and URL to the server, which
responds with the (possibly cached) page situation
references. A user can view situation references in a blog
post, as in the example, or in other web pages such as news
articles.
The back-end system uses manually created situation model
types (scripts) and currently supports kidnappings, legal
trials and corporate acquisitions each of which has multiple
possible outcomes and 8-12 possible events. The system Figure 4. Viewing an article for the selected situation
runs daily to retrieve news articles from several news web event.
sites via RSS feeds and store them in a Lucene index [5]. It
then queries the database for new articles with keywords Other work has focused on finding news articles relevant to
associated with the situation types it supports and reads blog posts by identifying and searching for important words
through the returned articles to create and extend situation in the post [8]. Though this approach doesn’t provide a
models instances of these types. These instances each high-level storyline view of a news situation that organizes
cover a specific situation such as the kidnapping of the relevant news articles, it benefits from being domain-
journalist above and include information from a few independent.
articles, up to several hundred if they are well-publicized.
Further discussion of Brussell appears in [7].
Brussell uses GATE [3], a standard open-source
information extraction system to extract situation REFERENCES
information including event references, dates and locations, 1. Akshay J., Finin, T., and Nirenburg, S. SemNews: A
and entity information such as person names and Semantic News Framework. 2006. Proceedings of the
occupations or organization names and nationalities. Twenty-First National Conference on Artificial
Extracting this information allows references such as "the Intelligence (AAAI-06), February 2006.
British journalist abducted last year" to be resolved to a
particular kidnapping situation instance. In fact, the same
2. Cayzer, S. Semantic Blogging: Spreading the Semantic
Web Meme. Proceedings of XML Europe 2004,
mechanism used for extracting information is used to
Amsterdam, Netherlands, April 2004.
identify situation references in page text, and in analyzing
news articles the system caches the textual references for all 3. Cunningham, H., Maynard, D., Bontcheva, K., Tablan,
of articles it processes. V. GATE: A Framework and Graphical Development
Environment for Robust NLP Tools and Applications.
RELATED WORK Proceedings of the 40th Anniversary Meeting of the
Several semantic-blogging systems and been developed to Association for Computational Linguistics (ACL'02).
support blog authoring for the Semantic Web. An extension Philadelphia, PA, USA, July 2002.
to the RDF-based personal information management system
Haystack supports right-clicking on any web page to create 4. Dzbor, M., Domingue, J., and Motta, E. Magpie:
a blog post with a link to the page [6]. Like Brussell, it also Towards a Semantic Web Browser. ISWC 2003, Second
improves the experience for reading blogs through International Semantic Web Conference, Sanibel Island,
structured views, including a graph of the comments made FL, USA, October 2003.
about a blog post. Unlike Brussell, it doesn’t offer content- 5. Lucene. http://lucene.apache.org/java/docs/
specific views of the news referenced in blog posts,
6. Karger, D. R., Quan, D. What would it mean to blog on
however. The SWAD-E project represents the
the semantic web? Journal of Web Semantics. 3(2-3):
bibliographic entries referenced within blog posts and
147-157. 2005.
provides content-specific affordances for interacting with
this information [2]. Another approach is taken by the 7. Wagner, E. J., Liu, J., Birnbaum, L., Forbus, K. D. Rich
Magpie Semantic Web browsing system, which allows rich Interfaces for Reading News on the Web. Proceedings
interaction with references to entities within web pages [4]. of the 2009 International Conference on Intelligent User
Neither system models news events and situations and how Interfaces, Sanibel Island, FL, USA, February 2009.
these situations unfold, however. 8. Ikeda, D., Fujiki, T., Okumura, M. Automatically
Some systems extract formal knowledge from news to Linking News Articles to Blog Entries. AAAI Spring
populate the Semantic Web such as SemNews, which Symposium: Computational Approaches to Analyzing
processes news retrieved via RSS feeds [1]. Unlike Weblogs, Palo Alto, CA, USA, 2006.
Brussell, however, its emphasis is generating
representations in the form of RDF triples rather than
presenting structured views to the user.
3