Báo cáo khoa học: "Learning to “Read Between the Lines” using Bayesian Logic Programs" pptx

... performance to theirs. Therefore, we compared to the following methods: • Logical Deduction: This method forward chains on the extracted facts using the first- order rules learned by LIME to infer ... Computational Linguistics Learning to “Read Between the Lines” using Bayesian Logic Programs Sindhu Raghavan Raymond J. Mooney Hyeonseo Ku Department of Computer S...

Ngày tải lên: 30/03/2014, 17:20

... features to determine the quality of the match while the the edge matching functions take into account the types of relations and the difference in lengths between the aligned paths. Following the ... opted to simply divide the total alignment score by the number of nodes in the in- structor answer. The second transformation scales the node match- ing score by mu...

Ngày tải lên: 07/03/2014, 22:20

... the goal is to find a hypothesis, H, given examples, E, and background knowl- edge, B, such that the hypothesis along with the background knowledge logically implies the examples (Muggleton, ... specifying the properties of the rules have to be given by the user. A modeh declaration specifies the head of the rules, while modeb declarations specify what the bod-...

Ngày tải lên: 17/03/2014, 07:20

... of pareto-optimal points. A hypothesis is weakly pareto-optimal if there is no other hypothesis that improves all the metrics; a hypothesis is pareto- optimal if there is no other hypothesis that improves at ... instance the pareto- optimal point (0.4,0.7). There is no other point with either (metric1 > 0.4 and metric2 ≥ 0.7), or (met- ric1 ≥ 0.4 and metric2 > 0.7). On the other h...

Ngày tải lên: 19/02/2014, 19:20

... phrase as query to a search engine to retrieve snippets, and generate features for each tokens in the same way as done in the training phase. We then use the trained model to tag the snippets, ... snippets and identify the translations of the given term. The system can potentially be used to assist translators to find the most common translation for a given term...

Ngày tải lên: 19/02/2014, 19:20

... shows that when the word above occurs, our policy prefers to go to the north of the target landmark. The bottom row shows the weights of egocentric (absolute) spatial features. The bottom left figure ... us to the next utterance to interpret, and moves our location to the target of the action. If a is the null action, s = (u i+1 , l, c), otherwise s  = (u i+1 , l...

Ngày tải lên: 20/02/2014, 04:20

... harder to detect than others, the system’s classification accuracy depends largely on the user’s barrier. To prevent classification accuracy distorting the user evaluations, we asked the subjects to ... are sent to an utterance classifier. The category assigned by this classifier is returned to the dialogue manager and triggers a transition to the next node in the dial...

Ngày tải lên: 20/02/2014, 12:20

... within the table. Fur- thermore, the two columns within the table appear right next to each other there are no blank vlines separating the two columns. Worse still, some words from the first ... the table are then grouped together to form the columns. 4.2.3 Row The deterministic algorithm to recognize table row is similar to the recognition algorithm of the...

Ngày tải lên: 20/02/2014, 19:20

Ngày tải lên: 22/02/2014, 03:20

... only the most-frequent rule is kept, and its count is the to- tal count of all the rules. This means that there is a one -to- one correspondence between the rules ex- tracted by fuzzy tree -to- tree ... other, rather than multiple restruc- turings of the training trees. 4.3 Rule usage We then took a closer look at the behavior of the string -to- string and fuzzy tree -to- tr...

Ngày tải lên: 07/03/2014, 22:20