TY - RPRT A1 - Lenhard, Wolfgang T1 - Bridging the Gap to Natural Language: A Review on Intelligent Tutoring Systems based on Latent Semantic Analysis N2 - One of the major drawbacks in the implementation of intelligent tutoring systems is the limited capacity to process natural language and to automatically deal with unexpected or unknown vocabulary. Latent Semantic Analysis (LSA) is a statistical technique of automatic language processing, which can attenuate the “language barrier” between humans and tutoring systems. LSA-based intelligent tutoring systems address the goals of modelling human tutoring dialogues (AutoTutor), enhancing text comprehension and summarisation skills (State-The-Essence, Summary Street®, conText, Apex), training of comprehension strategies (iStart, a French system in development) and improving story and essay writing (Write To Learn, Select-a-Kibitzer, StoryStation). The systems are reviewed concerning their efficacy in modelling skilled human tutors and regarding their effects on the learner. KW - Intelligentes Tutorsystem KW - Schreib- und Lesefähigkeit KW - Sprachverarbeitung KW - Automatische Sprachanalyse KW - intelligente tutorielle Systeme KW - Latente Semantische Analyse KW - Schreibfertigkeiten KW - intelligent tutoring systems KW - latent semantic analysis KW - writing-skills Y1 - 2008 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-27980 ER -