A collocation is a habitual word combination. Collocational knowledge is essential for many tasks in natural language processing. We present a method for extracting collocations from text corpora. By comparison with the SUSANNE corpus, we show that both high precision and broad coverage can be achieved with our method. Finally, we describe an application of the automatically extracted collocati...

Despite the big amount of general language dictionaries in electronic form, those coming from specialised areas are still “under construction”. There are two main reasons for this: firstly, the need for these dictionaries was/is less essential than the need for general language dictionaries, since these were/are aiming mainly to specialists, and secondly many specialised areas are changing over...

This paper proposes a bidirctional hierarchical clustering algorithm for simultaneously clustering words of different parts of speech based on collocations. The algorithm is composed of cycles of two kinds of alternate clustering processes. We construct an objective function based on Minimum Description Length. To. partly solve the problem caused by sparse data two concepts of collocational deg...

The goal of this paper is to test productive collocational competence German as a foreign language students at the University Debrecen. aims analyze compensational strategies that are used by in their texts. In first part we take look terms collocation and second show results written test.

The paper aims at serving two goals. Firstly, attention will be drawn to a class of collocations that, to our knowledge at least, has gone unnoticed so far. Secondly, it will be argued that brute force techniques as they are employed by certain automated collocation search strategies using a quantitative technique plus a mechanism to filter out unwanted combinations is a guaranteed way to miss ...