In this work we hypothesize that giving students a reflective reading after spoken dialog tutoring in qualitative physics will improve learning. The reading is designed to help students compare similar aspects of previously tutored problems, and to abstract their commonalities. We also hypothesize that student motivation will affect how well the text is processed, and so influence learning. We find that the beneficial effects of the reflective text significantly interact with motivation, such that moderately motivated students learn significantly more from the reflective text than from a non-reflective control text. More poorly or highly motivated students did not benefit from reflective text. These results demonstrate that implicit reflection can improve learning after dialog tutoring with a qualitative physics tutor. They further demonstrate that this result can be obtained with a reflective/abstractive text without recourse to dialog, and that the effectiveness of the text is sensitive to the motivation level of the student.

Question generation tools can be used to extract a question-answer database from text articles. We investigate how suitable this technique is for giving domain-specific knowledge to conversational characters. We tested these characters by collecting questions and answers from naive participants, running the questions through the character, and comparing the system responses to the participant answers. Characters gave a full or partial answer to 53% of the user questions which had an answer available in the source text, and 43% of all questions asked. Performance was better for questions asked after the user had read the source text, and also varied by question type: the best results were answers to who questions, while answers to yes/no questions were among the poorer performers. The results show that question generation is a promising method for creating a question answering conversational character from an existing text.

This paper presents an automatic extraction and classification approach of opinions in texts. Therefore, we propose a similarity measurement calculating semantically similarities between a word and predefined subgroups of seed words. We have evaluated our approach on the semantic evaluation company “SemEval 2007” corpus, and we obtained promising results: the best value of Precision, 62%; and F1, 61%; as an improvement of 20 % compared to the participant systems.

Tekkotsu is a free, open source software framework for high-level robot programming. We describe enhancements to Tekkotsu's navigation component, the Pilot, to incorporate a particle filter for localization and an RRT-based path planner for obstacle avoidance. This allows us to largely automate the robot's navigation behavior using a combination of odometry and landmark-based localization. Beginning robot programmers need only indicate a destination in Tekkotsu's world map and the Pilot will take the robot there. The software has been tested both in simulation and on Calliope, a new educational robot developed in the Tekkotsu lab in collaboration with RoPro Design, Inc..

In this paper, we present a Personalized Intelligent Tutoring System that uses Reinforcement Learning techniques to implicitly learn teaching rules and provide instructions to students based on their needs. The system works on coarsely labeled data with minimum expert knowledge to ease extension to newer domains.

We define a probe to be a single action sequence computedgreedily from a given state that either terminates in the goalor fails. We show that by designing these probes carefullyusing a number of existing and new polynomial techniquessuch as helpful actions, landmarks, commitments, and con-sistent subgoals, a single probe from the initial state solvesby itself 683 out of 980 problems from previous IPCs, a num-ber that compares well with the 627 problems solved by FFin EHC mode, with similar times and plan lengths. We alsoshow that by launching one probe from each expanded statein a standard greedy best first search informed by the addi-tive heuristic, the number of problems solved jumps to 900(92%), as opposed to FF that solves 827 problems (84%),and LAMA that solves 879 (89%). The success of probessuggests that many domains can be solved easily once a suit-able serialization of the landmarks is found, an observationthat may open new connections between recent work in plan-ning and more classical work concerning goal serializationand problem decomposition in planning and search.

Research on automated essay scoring (AES) indicates that computer-generated essay ratings are comparable to human ratings. However, despite investigations into the accuracy and reliability of AES scores, less attention has been paid to the feedback delivered to the students. This paper presents a method developers can use to quickly evaluate the usability of an automated feedback system prior to testing with students. Using this method, researchers evaluated the feedback provided by the Writing-Pal, an intelligent tutor for writing strategies. Lessons learned and potential for future research are discussed.

Past research has shown that dictionaries and glosses can be beneficial in computer assisted language learning, particularly in vocabulary learning. We propose that L2 vocabulary learners can benefit from the use of a dictionary whose definitions are sensitive to the provided reading context, and that advances in the natural language processing task of word sense disambiguation can be used to automatically order the definitions of such a dictionary. An in-vivo study was conducted with ESL students to investigate the effect that the order of definitions has on vocabulary learning using REAP, a computer based vocabulary tutor. Our results showed that students benefited from having the algorithmically determined best definitions listed at the top of the definition list. Furthermore, our results suggest that word sense disambiguation may currently be good enough for use in intelligent language tutoring environments.

Conformant planning can be formulated as a path-finding problem in belief space where the two main challenges are the heuristics to guide the search, and the representation and update of beliefs. In the translation-based approach recently introduced by Palacios and Geffner, the two aspects are handled together by translating conformant problems into classical ones that are solved with classical planners. While competitive with state-of-the-art methods, the translation-based approach runs however into three difficulties. First, complete translations are expensive for problems with high width; second, incomplete translations can generate infinite heuristic values for problems that are solvable; and third, aspects that are specific to the conformant setting, such as the cardinality of beliefs, are not accounted for. In this work, we build on the translation-based approach but not for solving conformant problems with a classical planner but for deriving heuristics and computing beliefs in the context of a standard belief-space planner. For this, a novel translation KSi is introduced that is always complete, but which is sound for problems with width bounded by i. A new conformant planner, called T1, builds then on this translation for i=1, extending the heuristic that results with a second heuristic obtained from invariant "oneof expressions". A number of experiments is performed to compare T1 with state-of-the-art conformant planners.

In this paper, we present a new algorithm that integrates recent advances in solving continuous bandit problems with sample-based rollout methods for planning in Markov Decision Processes (MDPs). Our algorithm, Hierarchical Optimistic Optimization applied to Trees (HOOT) addresses planning in continuous-action MDPs. Empirical results are given that show that the performance of our algorithm meets or exceeds that of a similar discrete action planner by eliminating the problem of manual discretization of the action space.