TY  - JOUR
T1  - Map? or List?based Recommender Agents? Does the Map Metaphor Fulfill its Promise?
JF  - Information Visualization
Y1  - 2016
A1  - Reitsma,Reindert
A1  - Hsieh,Ping-Hung
A1  - Diekema,Anne
A1  - Robson,Robby
A1  - Zarsky,Malinda
KW  - BIS
KW  - Supply Chain
AB  - We present a spatialization of digital library content based on item similarity and an experiment which compares the performance of this spatialization relative to a simple list-based display. Items in the library are K-12 science and engineering learning resources. Spatialization and visualization are accomplished through 2D interactive Sammon mapping of pairwise item similarity scores based on the joint occurrence of word bigrams. The 65 science teachers participating in the experiment were asked to search the library for curricular items they would consider using in conducting one or more teaching assignments. Results indicate that whereas the spatializations adequately capture the salient features of the library’s content and teachers actively use them, item retrieval rates, task-completion time and perceived utility do not significantly differ from the semantically poorer but easier to comprehend and navigate list-based representations. These results put into question the usefulness of the rapidly increasing supply of information spatializations.
VL  - 16
UR  - http://journals.sagepub.com/doi/abs/10.1177/1473871616669193
U2  - a
U4  - 127142455296
ID  - 127142455296
ER  - 

TY  - CONF
T1  - Estimation and Visualization of Digital Library Content Similarities
T2  - Intern. Conf. on Inf. Systems (ICIS) 2015
Y1  - 2015
A1  - Reitsma,Reindert
A1  - Hsieh,Ping-Hung
A1  - Robson,Robby
KW  - BIS
KW  - Supply Chain
AB  - We report on a process for similarity estimation and two-dimensional mapping of lesson materials stored in a Web-based K12 Science, Technology, Engineering and Mathematics (STEM) digital library. The process starts with automated removal of all information which should not be included in the similarity estimations followed by automated indexing. Similarity estimation itself is conducted through a natural language processing algorithm which heavily relies on bigrams. The resulting similarities are then used to compute a Sammon-map; i.e., a projection in n dimensions, the item-to-item distances of which best reflect the input similarities. In this paper we concentrate on specification and validation of this process. The similarity results show almost 100% precision-by-rank in the top three to five ranks. Sammon mapping in two dimensions corresponds well with the digital library‘s table of content.
JA  - Intern. Conf. on Inf. Systems (ICIS) 2015
U2  - b
U4  - 105616928768
ID  - 105616928768
ER  -