| Title | Representing Uncertainty with a New Type of Stochastic Neural Networks |
| Author | *Jumpol Polvichai, Surapont Toomnark (King Mongkut's University of Technology Thonburi, Thailand) |
| Page | pp. 609 - 612 |
| Keyword | Stochastic Neural Networks, Genetic Algorithms, Probability Modeling, Uncertainty |
| Abstract | Many interesting complex systems are stochastic. In order to model such complex systems, much ongoing research is looking at how to precisely model uncertainty in performance. In this paper, we proposed a novel type of stochastic neural network (SNN), in which dynamic features are added to the input layer allowing any non-deterministic system to be modeled. The SNNs capture randomness from the additional input nodes fed with internal random signals. These random signals, combined with weights between the additional nodes and the hidden nodes, allow stochastic output even though the network is deterministic. To validate this approach, a preliminary experiment was performed. To show the SNN's basic ability to represent uncertainty, a SNN model is trained to represent a model of beta distribution. Experiments verify the basic feasibility of the approach. |
| Title | Automatic Evaluation of Question Answering System based on BE Method |
| Author | *Akiko Yamamoto, Junichi Fukumoto (Ritsumeikan University, Japan) |
| Page | pp. 613 - 616 |
| Keyword | question answering, automatic evaluation, basic element, Pearson's correlation |
| Abstract | In this paper, we describe automatic evaluation method for question answering in natural language. This method is based on BEs (Basic Elements) originally proposed by Hovy et. al. for automatic evaluation of document summaries. We applied BE method for evaluation of question answering with comparison between BEs of system answer and BEs of correct answers. According to the experiments using QAC4 test set, we have proved that BE method has some correlation with human evaluation. |
| Title | Question Answering System beyond Factoid Type Questions |
| Author | *Satoshi Nakakura, Junichi Fukumoto (Ritsumeikan University, Japan) |
| Page | pp. 617 - 620 |
| Keyword | question answering, named entity extraction, non-factoid question, answer extraction, RST |
| Abstract | In this paper, we describe answer extraction method for non-factoid questions. We classified non-factoid type questions into three types: why type, definition type and how type. We analyzed each type of questions and developed answer extraction patterns for these types of questions. For each question type, we have expanded question analysis modules to determine non-factoid question types and developed answer extraction modules based on the analysis of answer expression patterns in large document set. For evaluation, we used 104 questions which are mainly developed at Question Answering evaluation workshop (NTCIR6-QAC4). |