No abstract

No abstract

No abstract

No abstract

Although technology for automatic grading of multiple choice exams has existed for several decades, it is not yet as widely available or affordable as it should be. The main reasons preventing this adoption are the cost and the complexity of the set-up procedures. In this article, Eyegrade, a system for automatic grading of multiple choice exams is presented. Whilst most current solutions are based on expensive scanners, Eyegrade offers a truly low-cost solution requiring only a regular off-the-shelf webcam. Additionally, Eyegrade performs both mark recognition as well as optical character recognition (OCR) of hand-written student identification numbers, which avoids the use of bubbles in the answer sheet. When compared to similar webcam-based systems, the user interface in Eyegrade has been designed to provide a more efficient and error-free data collection procedure. The tool has been validated with a set of experiments that show the ease of use (both set-up and operation), the reduction in grading time, and an increase in the reliability of the results when compared with conventional, more expensive systems.

No abstract

No abstract

This paper presents a hash and a canonicalization algorithm for Notation 3 (N3) and Resource Description Framework (RDF) graphs. The hash algorithm produces, given a graph, a hash value such that the same value would be obtained from any other equivalent graph. Contrary to previous related work, it is well-suited for graphs with blank nodes, variables and subgraphs. The canonicalization algorithm outputs a canonical serialization of a given graph (i.e. a canonical representative of the set of all the graphs that are equivalent to it). Potential applications of these algorithms include, among others, checking graphs for identity, computing differences between graphs and graph synchronization. The former could be especially useful for crawlers that gather RDF/N3 data from the Web, to avoid processing several times graphs that are equivalent. Both algorithms have been evaluated on a big dataset, with more than 29 million triples and several millions of subgraphs and variables.