With the growing number of document sets accessible online, tracking their evolution over time (story tracking) became an increasingly interesting problem. In this paper we propose a story tracking method based on the dynamics of keyword-association graphs. We create a graph representation of the s tory evolution that we call story graphs, and investigate how graph structure can be used for detecting and discovering new developments in the story. First we investigate the possibly interesting graph properties for development detection. We continue by investigating how graph structure can be linked to the sentences representing developments. For this we create an evaluation framework which bridges the gap between temporal text mining patterns and sentences. We apply this framework to evaluate our method against other temporal text mining methods. Our experiments show that story graphs perform at similar levels overall, but provide distinctive advantages in some settings.

The Internet has become for many the most important medium for staying informed about current news events. Some events cause heightened interest on a topic, which in turn yields a higher frequency of the search queries related to it. These queries ar

Many document collections are by nature dynamic, evolving as the topics or events they describe change. The goal of temporal text mining is to discover bursty patterns and to identify and highlight these changes to better enable readers to track stories. Here, we focus on the news domain, where the changes revolve around novel, previously unpublished, “facts” that have an effect on the story developments. However, despite intense research activities on bursty patterns, a lack of common procedures today makes it impossible to compare methods in a principled way. To close this gap, we (a) investigate how different temporal text mining methods discover novel facts and (b) present an evaluation framework for methods assessment, consisting of a set of procedures and metrics for cross-evaluating models. Bursty patterns are transformed into queries for sentence retrieval, either with or without taking into account internal pattern structure, and these sentences are compared with a set of editor-selected ground-truth reference sentences. Our experiments on different classes of temporal text mining show that different methods perform at similar levels overall, but provide distinctive advantages in some settings. The experiments also demonstrate the benefits of using patterns' internal structure for query generation.

Using data collections available on the Internet has for many people became the main medium for staying informed about the world. Many of these collections are in nature dynamic, evolving as the subjects they describe change. The goal of different research areas is to identify and highlight these changes to better enable readers to track stories. In this work we restrict ourselves to news collections and investigate "real-life" effectiveness and usability of temporal text mining (TTM) story tracking methods. We propose a new story tracking method and build a tool to support it. Additionally, we investigate the effectiveness and usability of story tracking methods and define a new frameworks for automatic and user oriented evaluation. We built methods and tools which allow for understanding, discovery, and search through user interaction. Although there are many TTM methods developed there is a lack of common evaluation procedure. Therefore, we propose an evaluation framework for measuring how different TTM methods discover novel "facts". Apart from the automatic evaluation we are interested in how can users interact with pattens and learn about the underlying subjects of the story they track. For this purpose we propose a user testing environment that measures speed and accuracy in which users can use story tracking methods to discover predefined sets of ground-truth sentences.

A query burst is a period of heightened interest of users on a topic which yields a higher frequency of the search queries related to it. In this paper we examine the behavior of search engine users during a query burst, compared to before and after this period. The purpose of this study is to get insights about how search engines and content providers should respond to a query burst. We analyze one year of web-search logs, looking at query bursts from two perspectives. First, we adopt the user's perspective describing changes in user's effort and interest while searching. Second, we look at the burst from the general content providers' view, answering the question of under which conditions a content provider should ``ride'' a wave of increased interest to obtain a significant share of clicks.

Rich information spaces (like the Web or scientific publications) are full of "stories": sets of statements that evolve over time, manifested as, for example, collections of newspaper articles reporting events relating to an evolving crime investigation, sets of news articles and blog posts accompanying the development of a political election campaign, or sequences of scientific papers on a topic. In this paper, we propose a method and a visualisation tool for mapping and interacting with such stories. In contrast to existing approaches, our method concentrates on relational information and on local patterns rather than on the occurrence of individual concepts and global models. In addition, we present an evaluation framework. A real-life case study is used to illustrate and evaluate the method and tool.