An agent-based methodology for analyzing and visualizing educational assessment data

We examine data collected from on-line assessments of the numeracy and literacy skills of young students in order to construct probabilistic agent-based controllers. We demonstrate the value of this methodology as an effective means for both analyzing and visualizing aspects of large data sets that are difficult to capture with traditional equation-based statistics and static graphics. Categories and Subject Descriptors I.2.11 [Artificial Intelligence]: Distributed Artificial Intelligence; K.3.1 [Computers and Education]: Computer Uses in Education General Terms Design, Human Factors Keywords learning, training, education, simulation

• Type: Conference
• Year: 2007
• Where: ATAL
• Authors: Elizabeth Sklar, Jordan Salvit, Christopher Camacho, William Liu
• Keywords: Artificial Intelligence, ATAL 2007, Probabilistic Agent-based Controllers, Traditional Equation-based Statistics