| 1. |
- Borg, Markus, et al.
(author)
-
Time extraction from real-time generated football reports
- 2007
-
In: [Host publication title missing]. - 9789985405130 ; , s. 37-43
-
Conference paper (peer-reviewed)abstract
- This paper describes a system to extract events and time information from football match reports generated through minute-by-minute reporting. We describe a method that uses regular expressions to find the events and divides them into different types to determine in which order they occurred. In addition, our system detects time expressions and we present a way to structure the collected data using XML.
|
|
| 2. |
- Granfeldt, Jonas, et al.
(author)
-
Evaluating stages of development in second language French: A machine-learning approach
- 2007
-
In: NODALIDA 2007 PROCEEDINGS. - 9789985405147
-
Conference paper (peer-reviewed)abstract
- This paper describes a system to define and evaluate development stages in second language French. The identification of such stages can be formulated as determining the frequency of some lexical and grammatical features in the learners’ production and how they vary over time. The problems in this procedure are threefold: identify the relevant features, decide on cutoff points for the stages, and evaluate the degree of success of the model. The system addresses these three problems. It consists of a morphosyntactic analyzer called Direkt Profil and a machine-learning module connected to it. We first describe the usefulness and rationale behind its development. We then present the corpus we used to develop the analyzer. Finally, we present new and substantially improved results on training machine-learning classifiers compared to previous experiments (Granfeldt et al., 2006). We also introduce a method to select attributes in order to identify the most relevant grammatical features.
|
|
| 3. |
- Johansson, Richard, et al.
(author)
-
Extended constituent-to-dependency conversion for English
- 2007
-
In: NODALIDA 2007 Proceedings. - 9789985405147 ; , s. 105-112
-
Conference paper (peer-reviewed)abstract
- We describe a new method to convert English constituent trees using the Penn Treebank annotation style into dependency trees. The new format was inspired by annotation practices used in other dependency treebanks with the intention to produce a better interface to further semantic processing than existing methods. In particular, we used a richer set of edge labels and introduced links to handle long-distance phenomena such as wh-movement and topicalization. The resulting trees generally have a more complex dependency structure. For example, 6% of the trees contain at least one nonprojective link, which is difficult for many parsing algorithms. As can be expected, the more complex structure and the enriched set of edge labels make the trees more difficult to predict, and we observed a decrease in parsing accuracy when applying two dependency parsers to the new corpus. However, the richer information contained in the new trees resulted in a 23% error reduction in a baseline FrameNet semantic role labeler that relied on dependency arc labels only.
|
|
