Against the background of a broad range of language features that are identified as characteristic of specific language impairment (SLI), some researchers have identified a narrower set of clinical markers considered the hallmark of SLI. However, comparisons with language impairments that fall outside the criteria for SLI are limited. This thesis is concerned with determining which language features, if any, are capable of differentiating children with SLI from children with non-specific language impairment (NLI). Conversation and oral narrative language samples were collected from seventy five children aged 2 ½ to 6 years comprising four research groups: 21 participants with SLI, 13 participants with NLI, 21 age-matched participants with normally developing language (AM) and 20 younger language-matched participants with normally developing language (LM). Matching for group comparisons required that the SLI and NLI groups had similar levels of language ability on a standardised assessment and mean length of utterance (MLU), which reduced the SLI group to 15 participants for these comparisons. The LM group was also matched to the SLI and NLI groups on MLU. A wide range of language variables from the conversation and narrative samples were analysed, covering the domains of general sample measures, morphosyntactic accuracy and complexity, narrative structure and cohesion. The SLI and NLI groups performed similarly in all domains and could not be differentiated diagnostically on the measures examined. The most consistent group effects were for comparisons between the AM and LM groups, which demonstrated the effects of maturation and development. The language impairment (LI) and LM groups could not be differentiated on the majority of general language sample or morphosyntactic measures but the SLI group produced narratives that were structurally more complex and cohesive than the LM group. Language tasks varied in their effectiveness in differentiating groups. More consistent group effects for the grammatical accuracy measures were obtained from the conversations than the narratives, and from composite measures compared to individual morpheme measures. Targeted elicitation tasks were more effective than the conversations or narratives in producing consistent group effects for accuracy of individual verb tense morphemes. More consistent group effects for the narrative features were obtained from a wordless picture book than a single scene picture. A set of discriminant function analyses showed that LI was most effectively identified using a combination of key morphosyntactic measures from the conversations and key narrative feature measures from the two narratives. The results have implications for diagnostic practices, intervention practices and theoretical constructs and explanations of SLI and NLI. In particular, a broad, holistic view of LI is supported, as an impairment that impacts on all domains of language which interact with each other and must be considered collectively, rather than as individual, splintered skills.

This report explores of the feasibility of using pattern matching for implementing a robust Natural Language Interface (NLI) over a limited Application Programming Interface (API). Because APIs are used to such a great extent today and often in mobile applications, it becomes more important to find simple ways of making them accessible to end users. A very intuitive way to access information via an API is using natural language. Therefore, this study first explores the possibility of building a corpus of the most common phrases used for a particular API. It is then explored how those phrases adhere to patterns, and how these patterns can be used to extract meaning from a phrase. Finally it evaluates an implementation of an NLI using pattern matching system based on the patterns. The result of the building of the corpus shows that although the amount of unique phrases used with our API seems to increase quite steadily, the amount of patterns those phrases follow converges to a constant quickly. This implies that it is possible to use these patterns to create an NLI that is robust enough to query an API effectively. The evaluation of the pattern matching system indicates that this technique can be used to successfully extract information from a phrase if its pattern is known by the system. / Den här rapporten utforskar hur genomförbart det är att använda mönstermatchning för att implementera ett robust användargränssnitt för styrning med naturligt språk (Natural Language Interface, NLI) över en begränsad Application Programming Interface (API). Eftersom APIer används i stor utsträckning idag, ofta i mobila applikationer, har det blivit allt mer viktigt att hitta sätt att göra dem ännu mer tillgängliga för slutanvändare. Ett mycket intuitivt sätt att komma åt information är med hjälp av naturligt språk via en API. I den här rapporten redogörs först för möjligheten att bygga ett korpus för en viss API and att skapa mönster för mönstermatchning på det korpuset. Därefter utvärderas en implementation av ett NLI som bygger på mönstermatchning med hjälp av korpuset. Resultatet av korpusuppbyggnaden visar att trots att antalet unika fraser som används för vårt API ökar ganska stadigt, så konvergerar antalat mönster på de fraserna relativt snabbt mot en konstant. Detta antyder att det är mycket möjligt att använda desssa mönster för att skapa en NLI som är robust nog för en API. Utvärderingen av implementationen av mönstermatchingssystemet antyder att tekniken kan användas för att framgångsrikt extrahera information från fraser om mönstret frasen följer finns i systemet.

Native language identification (NLI) is a classification task which identifies the mother tongue of a language learner based on spoken or written material. The task gained popularity when it was featured in the 2017 BEA-12-workshop and since then many applications have been successfully found for NLI - ranging from language learning to authorship identification and forensic science. While a considerable amount of research has already been done in this area, we introduce a novel approach of incorporating syntactic information into the implementation of a BERT-based NLI model. In addition, we train separate models to test whether erroneous input sequences perform better than corrected sequences. To answer these questions we carry out both a quantitative and qualitative analysis. In addition, we test our idea of implementing a BERT-based GEC model to supply more training data to our NLI model without the need for manual annotation. Our results suggest that our models do not outperform the SVM baseline, but we attribute this result to the lack of training data in our dataset, as transformer-based architectures like BERT need huge amounts of data to be successfully fine-tuned. In turn, simple linear models like SVM perform well on small amounts of data. We also find that erroneous structures in data come useful when combined with syntactic information but neither boosts the performance of NLI model separately. Furthermore, our implemented GEC system performs well enough to produce more data for our NLI models, as their scores increase after implementing the additional data, resulting from our second experiment. We believe that our proposed architecture is potentially suitable for the NLI task if we incorporate extensions which we suggest in the conclusion section.