Colin DeLong

Web Mining For Self-directed E-learning

Self-directed e-learning focuses on the independent learner, one who engages in education at his own pace, free from curricular obligation. A number of tools, some purposefully and others serendipitously, have become key enablers of this learning paradigm. For example, tools such a Google Scholar, CiteSeer Research Index, etc. make it possible to do literature search without stepping out of one’s room. Due to the same technologies which helped make self-directed e-learning possible in the first place, these tools are in danger of delivering diminishing returns as micro-learning, lifelong education, and continuous education become the norm in our Information Age. Web Mining, however, may potentially offer a solution to this issue. In this chapter, we investigate specific examples of selfdirected e-learning and how their functionality and utility can be improved through the use of Web Mining technology, techniques, and practices. Our work demonstrates the usefulness of Web Mining as it applies to self-directed elearning and the need to map implicit relationships in learner behaviour, usage, and context.

Player and Team Performance in Everquest II and Halo 3

The market for video games has skyrocketed over the past decade. In the United States alone, the video game industry in 2009 generated almost US

TeamSkill and the NBA: applying lessons from virtual worlds to the real-world

20 billion in sales. Furthermore, according to Lenhart et al. (2008), an estimated 97% of the teenage population and 53% of the adult population are regular game players. Massively multiplayer online games (MMOGs) have become increasingly popular and amassed communities comprised of over 47 million subscribers by the year 2008. MMOGs are online spaces providing users with comprehensive virtual universes, each with its own unique context and mechanics. They range from the fantastical world of elves, dwarfs, and humans to space faring corporations and mirrors of our world. Large numbers of users interact and role-play via in-game mechanics.

Concept-Aware Ranking: Teaching an Old Graph New Moves

In this paper, we build on our previous work by evaluating several approaches for assessing the skill of players and teams on the basis of both individual performance and group cohesion, or “team chemistry”, using game data from the National Basketball Association (NBA). Previously developed for skill assessment in team-based multi-player video games (e.g., Halo 3), we find that group cohesion is a predictive feature in virtual and real-world team-based games, and that methods utilizing such features can often outperform the baseline in both contexts. Additionally, we observe a strong positive correlation between the predictive accuracy of our group cohesion-based approaches and the duration of playing time between a particular configuration of players on a team and their opponents, or “match-up” length.

An Exploratory Study of Player and Team Performance in Multiplayer First-Person-Shooter Games

In ranking algorithms for Web graphs, such as PageRank and HITS, the lack of attention to concepts/topics representing Web page content causes problems such as topic drift and mutually reinforcing relationships between hosts. This paper proposes a novel approach to expand the Web graph to incorporate conceptual information encoded by links (anchor text) between Web pages. Using Web graph link structure and conceptual information associated with each Web page (automatically extracted from anchor text of Minks), a new graph is defined where each node represents a unique pair of a Web page and concept associated with that Web page, and an edge represents an explicit or implicit link between two such nodes. This graph captures inter-concept relationships, which is then utilized by ranking algorithms. Our experimental results show that such an approach improves accuracy (e.g., first X precision) by retrieving links which are more authoritative given a users context

Social Topic Models for Community Extraction

In this paper, we report findings from an exploratory study of player and team performance in Halo 3, a popular First-Person-Shooter game developed by Bungie. In the study, we first analyze player and team statistics obtained from the 2008 and 2009 seasons for professional Halo 3 games in order to investigate the impact of change in team composition on player and team performance. We then examine the impact of past performance on future performance of players and teams. Performing a large-scale experiment on a real-world dataset, we observe that player and team performance can be predicted with fairly high accuracy and that information about change in team composition can further improve the prediction results.