In linguistics, words are usually considered to be composed of morphemes: units that carry meaning and are not further subdivisible. The task of this thesis is to create an automatic method for segmenting Czech words into morphemes, usable within the network of Czech derivational relations DeriNet. We created two different methods. The first one finds morpheme boundaries by differentiating words against their derivational parents, and transitively against their whole derivational family. It explicitly models morphophonological alternations and finds the best boundaries using maximum likelihood estimation. At worst, the results are slightly worse than the state of the art method Morfessor FlatCat, and they are significantly better in some settings. The second method is a neural network made to jointly predict segmentation and derivational parents, trained using the output of the first method and the derivational pairs from DeriNet. Our hypothesis that such joint training would increase the quality of the segmentation over training purely on the segmentation task seems to hold in some cases, but not in other. The neural model performs worse than the first one, possibly due to being trained on data which already contains some errors, multiplying them.

Morphological segmentation is segmentation of words into morphemes - smallest units carrying meaning. It is a low level Natural Language Processing task. Since morphological segmentation is sometimes used as method of preprocessing, achieving better results on this task may help NLP algorithms to better solve various problems, especially in scenarios involving small amount of data, and it may also also help the linguistic research. We propose a novel ensemble algorithm for morphological segmentation of Czech lemmas which makes use of the DeriNet derivation tree dataset. As a sideproduct we also created suggestions for improvements of the DeriNet dataset.

Automatic morpheme segmentation as a field has been dominated by unsupervised methods since its inception. Partly due to theoretical motivations, but also due to resource constraints. Given the success neural network methods have shown on a wide variety of field in later years, it would seem compelling to apply these methods to the morpheme segmentation field. This study explores the efficacy of modern neural networks, specifically convolutional neural networks and Bi-directional LSTM networks, on the morpheme segmentation task in a resource low setting to determine their viability as contenders with previous unsupervised, minimally supervised, and semi-supervised systems in the field. One architecture of each type is implemented and trained on a new gold standard data set and the results are compared to previously established methods. A qualitative error analysis of the architectures’ segmentations is also performed. The study demonstrates that a BLSTM system can be trained with minimal effort to produce a proof of concept solution at low levels of training data and suggests that BLSTM methods may be a fruitful direction for further research in this field.

Raw text data online has increased the need for designing artificial systems capable of processing raw data efficiently and at a low cost in the field of natural language processing (NLP). A well-developed morphological analysis is an important cornerstone of NLP, in particular when word look-up is an important stage of processing. Morphological analysis has many advantages, including reducing the number of word forms to be stored computationally, as well as being cost-efficient and time-efficient. NLP is relevant in the field of medicine, especially in automatic text analysis, which is a relatively young field in Swedish medical texts. Much of the stored information is highly unstructured and disorganized. Using raw corpora, this paper aims to contribute to automatic morphological segmentation by experimenting with state-of-art-tools for unsupervised and semi-supervised word segmentation of Swedish words in medical texts. The results show that a reasonable segmentation is more dependent on a high number of word types, rather than a special type of corpora. The results also show that semi-supervised word segmentation in the form of annotated training data greatly increases the performance. / Rå textdata online har ökat behovet för artificiella system som klarar av att processa rå data effektivt och till en låg kostnad inom språkteknologi (NLP). En välutvecklad morfologisk analys är en viktig hörnsten inom NLP, speciellt när ordprocessning är ett viktigt steg. Morfologisk analys har många fördelar, bland annat reducerar den antalet ordformer som ska lagras teknologiskt, samt så är det kostnadseffektivt och tidseffektivt. NLP är av relevans för det medicinska ämnet, speciellt inom textanalys som är ett relativt ungt område inom svenska medicinska texter. Mycket av den lagrade informationen är väldigt ostrukturerat och oorganiserat. Genom att använda råa korpusar ämnar denna uppsats att bidra till automatisk morfologisk segmentering genom att experimentera med de för närvarande bästa verktygen för oövervakad och semi-övervakad ordsegmentering av svenska ord i medicinska texter. Resultaten visar att en acceptabel segmentering beror mer på ett högt antal ordtyper, och inte en speciell sorts korpus. Resultaten visar också att semi-övervakad ordsegmentering, dvs. annoterad träningsdata, ökar prestandan markant.