2014-09-09: DL2014 Doctoral Consortium

After exploring London on Sunday, I attended the first DL2014 session: the Doctoral Consortium. Held in the College Building at the City University London, the Doctoral Consortium offered early-career Ph.D. students the opportunity to present their research and academic plans and receive feedback from digital libraries professors and researchers.

Edie Rasmussen chaired the Doctoral Consortium. I was a presenter at the Doctoral Consortium in 2012 with Hany SalahEldeen, but I attended this year as a Ph.D. student observer.

Session I: User Interaction was chaired by José Borbinha. Hugo Huurdeman was first to present his work entitled "Adaptive Search Systems for Web archive research". His work focuses on information retrieval and discovery in the archives. He explained the challenge with searching not only across documents but also across time.

Georgina Hibberd presented her work entitled "Metaphors for discovery: how interfaces shape our relationship with library collections." Georgina is working on digitally modeling the physical inputs library users receive when interacting with books and physical library media to allow the same information to be available when interacting with digital representations of the collection. For example, how can we incorporate physical proximity and accidental discovery in the digital systems, or how can we demonstrate frequency of use that would previously be shown in the condition of a book's spine?

Yan Ru Guo presented her work entitled "Designing and Evaluating an Affective Information Literacy Game" in which she proposes serious games to help tertiary students in an effort to help their ability to perform searches and information discovery in digital environments.

After a break to re-caffeinate, Session II: Working with Digital Collections began. Dion Goh chaired the session. Vincent Barrallon presented his work entitled "Collaborative Construction of Updatable Digital Critical Editions: A Generic Approach." This work aims to establish an updatable data structure to represent the collaborative flow of annotation, especially with respect to editorial efforts. He proposes using bidirectional graphs, triple graphs, or annotated graphs as representatives, and proposes methods of identifying graph similarity.

Hui Li finished the session with her presentation entitled "Social Network Extraction and Exploration of Historic Correspondences" in which she is working to use Named Entity Extraction to create a social network from digitized historical documents. Her effort utilizes topic modeling and event extraction to construct the network.

Due to a scheduling audible, lunch and Session III: Social Factors overlapped slightly. Ray Larson chaired this session, and Mat Kelly was able to attend after landing in LHR and navigating to our hotel. Maliheh Farrokhnia presented her work entitled "A request-based framework for FRBRoo graphical representation: With emphasis on heterogeneous cultural information needs." Her work takes user interests (through adaptive selection of target information) to present relational graphs of digital library content.

Abdulmumin Isah presented his work entitled "The Adoption and Usage of Digital Library Resources by Academic Staff in Nigerian Universities: A Case Study of University of Ilorin." His work highlights a developing country's use of digital resources in academia and cites factors influencing the success of digital libraries.

João Aguir Castro presented his work entitled "Multi-domain Research Data Description -- Fostering the participation of researchers in an ontology based data management environment." His work with Dendro uses metadata and ontologies to aid in long-term preservation of research data.

The last hour of the consortium was dedicated to an open mic session chaired by Cathy Marshall with the goal of having the student observers present their current work. I presented first and explained my work that aims to mitigate the impact of JavaScript on archived web pages. Mat went next and discussed his work about integrating public and private web archives with tools like WAIL and WARCreate.

Alexander Ororbia presented his work on using artificial intelligence and deep learning for error correcting crowd sourced data from scholarly texts. Md Arafat Sultan discussed his work on using natural language processing to detect similarity in text to identify text segments that adhered to set standards (e.g. educational standards). Kahyun Choi discussed her work on perceived mood in eastern music from the point of view of western listeners. Finally, Fawaz Alarfaj discussed his work using entity extraction, information retrieval, and natural language processing to identify experts within a specified field.

As usual, the Doctoral Consortium was full of interesting ideas, valuable recommendations, and highly motivated Ph.D. students. Tomorrow marks the official beginning of DL2014.


--Justin F. Brunelle