Báo cáo khoa học: "Detecting problematic turns in human-machine interactions: Rule-induction versus memory-based learning approaches" pot

... Detecting problematic turns in human-machine interactions: Rule-induction versus memory-based learning approaches Antal van den Bosch ILK / Comp. Ling. KUB, Tilburg The Netherlands antalb@kub.nl Emiel ... situations using machine learning techniques. For instance, Litman et al. (1999) and Walker et al. (2000a) use RIPPER (Cohen 1996) to classify problematic and unproblem...

Ngày tải lên: 23/03/2014, 19:20

... models are obtained using the minimum description length (MDL) principle. MDL selects an appropriate model by compar- ing potential candidates in terms of the cost of storing the model and ... The minimum description length principle is then used to produce a model and cost for storing the head noun instances from a training corpus at the relevant argument slots. Alternating sub- ....

Ngày tải lên: 31/03/2014, 04:20

... tagging. 4 Machine learning components in Argo In order to ensure flexibility in building workflows, we split the machine learning capability into three distinct processing components, namely feature ... visible in the middle of Figure 1. 122 (a) Training (b) Tagging Figure 2: Two generic workflows demonstrating the use of the Feature Generator component for (a) training and (b) tagg...

Ngày tải lên: 16/03/2014, 20:20

... to distin- guish the types depending on whether an information or information sharin~ interaction zs involved. C interaction is przmarily information seeking, although some sharing interaction ... which have more of ~ining function. Training in- volves more of information sharing, while service in- volves more of providing infornmtion requested by the user. 2. Information about t...

Ngày tải lên: 17/03/2014, 19:20

... be most effective in improving chunking results. Indeed, our experi- ments show that introducing morphological fea- tures improves chunking quality by as much as 3-point in F-measure when compared ... hints, which we exploit using an SVM learning method. The resulting method reaches perform- ance in Hebrew comparable to the best results published in English. 2 Previous Work T...

Ngày tải lên: 23/03/2014, 18:20

... sometimes elim- inate the original meaning by incorrectly removing important parts of sentences, be- cause trimming probabilities only depend on parents’ and daughters’ non-terminals in applied CFG ... to unsupervised learning to overcome the lack of training data. However their model also has the same problem. McDonald (McDonald, 2006) independently proposed a new machine learning appr...

Ngày tải lên: 20/02/2014, 12:20

... language learning capabilities, or do we solely use the input from the environment to find struc- ture in language? Nativists believe that infants have an innate capability for acquiring language. ... the plot in figure 5 that the system begins life with no word representations. At the beginning, the system hypothesises new word units from which it can begin to bootstrap its int...

Ngày tải lên: 22/02/2014, 02:20

... occur in the original bilingual parallel cor- pora used for phrase table extraction. Our hypothe- sis is that the increase in recall obtained from relaxed matches through semantic tags in place ... matching terms and the correct entail- ment decisions is less strong. In such framework, for instance, the full mapping of the hypothesis into the text is per se not sufficient to discrimina...

Ngày tải lên: 19/02/2014, 19:20

... occurring in a hand-cleaned sub-corpus, as well as linguistic intuition. Using this method, Kveain and Oliva (2002) report find- ing 2661 errors in the NEGRA corpus (containing 396,309 tokens). Interestingly, ... work Considering the significant effort that has been put into obtaining pos-tagged reference corpora in the past decade, there are surprisingly few pub- lications on the issue...

Ngày tải lên: 22/02/2014, 02:20

... use cen- tering theory (Kameyama, 1986) to determine how easily a noun phrase can be referred to in the follow- ing context. 2.2 Centering Theory Centering theory is an empirical sorting rule used ... 2004) 3. STR-CT : STR with the centering top feature explained in Section 2.3. 4. STR-CS : STR with the centering structure fea- ture explained in Section 2.4. 3.1 Setting We used 1451 tex...

Ngày tải lên: 17/03/2014, 04:20