The development and research of Artificial Intelligence have had a recent surge in recent years, which includes the medical field. Despite the new technology and tools available, the staff is still under a heavy workload. The goal of this thesis is to analyze the possibilities of a chatbot whose purpose is to assist the medical staff and provide safety for the patients by guaranteeing that they are being monitored. With the use of technologies such as Artificial Intelligence, Natural Language Processing, and Voice Over Internet Protocol, the chatbot can communicate with the patient. It will work as an assistant for the working staff and provide the information from the calls to the medical staff. With the answers provided from the call, the staff will not be needing to ask routine questions every time and can provide help more quickly. The chatbot is administrated through a web application where administrators can initiate calls and add patients to the database.

In this paper, Natural Language Processing and classification algorithms were used to create a program that automatically can tag different errands that are connected to Fortnox (an IT company based in Växjö) support service. Controlled experiments were conducted to find the best classification algorithm together with different Bag-of-Word pre-processing algorithms to find what was best suited for this problem. All data were provided by Fortnox and were manually labeled with tags connected to it as training and test data. The result of the final algorithm was 69.15% correctly/accurately predicted errands using all original data. When looking at the data that were incorrectly predicted a pattern was noticed where many errands have identical text attached to them. By removing the majority of these errands, the result was increased to 94.08%.

In this thesis, we show how a text-based Recommendation Systems can greatly benefit from neural statistical language models, more particularly BERT. We evaluate the framework on a digital and collaborative platform for radiologists, by automatically suggesting scientific papers from the medical database PubMed, to provide evidence in diagnostic radiology. The models use contextualized vectors to represent text, accounting for writing style, misspelling and jargon. By using pre-computed representations of text passages, we are able to use compute-heavy statistical language models in production environments, where supercomputers are not available during inference. The results suggest pre-computed embeddings are very effective when the texts came from the same domain, and less effective (but still useful) in capturing the interaction between clinical and scientific text. Nonetheless, the suggested solutions hold promises in this and other areas in medicine. Possibly, the results are transferable to other domains, such as processing of legal documents and patent search.

This thesis analyses the biggest reward-based crowdfunding platform in the Czech Repub- lic using textual tools on uniquely collected microdata. The research question investigates which of the attributes of project campaigns (including the language style of project de- scriptions) have a significant impact on successful funding. Empirical analysis combines results of Bayesian Model Averaging and logistic regression. Results reveal that firstly, language style of project descriptions does not possess any significant predictive power. Secondly, that utilization of a video, size of pledging goal, or the number of contributors have a significant effect on the campaign's success, which is in line with current literature. Thirdly, it has proven to be true that project categorization plays an important role as well. On the contrary, the findings do not imply any causal claims, such as whether those factors persuade contributors to donate money. JEL Classification G23 Keywords crowdfunding, rewards, success determinants, natural language processing Author's e-mail t.hudcova@gmail.com Supervisor's e-mail polakpet@gmail.com

The way people acquire knowledge has largely shifted from print to web resources. Meanwhile, search has become the main medium to access information. Amongst various search behaviors, exploratory search represents a learning process that involves complex cognitive activities and knowledge acquisition. Research on exploratory search studies on how to make search systems help people seek information and develop intellectual skills. This research focuses on information retrieval and aims to build an exploratory search system that shows higher clustering performance and diversified search results. In this study, a new language model that integrates the state-of-the-art vector language model (i.e., BERT) with human knowledge is built to better understand and organize search results. The clustering performance of the new model (i.e., RDF+BERT) was similar to the original model but slight improvement was observed with conversational texts compared to the pre-trained language model and an exploratory search baseline. With the addition of the enrichment phase of expanding search results to related documents, the novel system also can display more diverse search results.

With the booming development of information technology, text information is not only remained in paper-based forms, but also in digital forms which have been distributed all over internet. Massive information on the internet provides us so many options while at the same time makes it hard for us to choose which detail information we exactly need. The appearance of media monitoring is going to change the situation and help solve the problem. Meltwater group as a media monitoring company provides a service of tracking and sorting information to enterprises and help them to achieve business goals. These goals may include finding the best time or place to do business campaign and knowing the dynamic information about the competitors. There is a recommender system in Meltwater. When a query has been searched, the corresponding documents which are searched from the database will be presented. The problem for the system is that some of the documents have beenturned out to be misclassified and the correctness rate for the recommendation isnot that high. To help solve this problem and make the search better, this paper will introduce a new algorithm which is based on word embedding approach and users’ supervision. The background information of Meltwater group and its existing frame of recommender system will be specifically illustrated at the beginning of the paper. Followed by it will be the exploration of background methods which include LSA (Latent Semantic Analysis), Random Indexing and Word2vec. Besides, the necessary tools such as T-SNE, K-means clustering and hierarchy clustering will also be mentioned in this part. The data sets that are going to be used in this paper will be described after thepart of background methods. Information such as the introduction of the data and the dealing of it will be mentioned in a detail way. The description of the algorithm will appear in the middle of the paper with detail steps. Followed by it is the evaluation. The algorithm will be evaluated by using several different data sets and the confusion matrix will be used as a means of measurement. Finally, a summary of the method as well as future suggestions will be made at the end of the paper.

It is of utmost importance for software developers to develop a product that is functional for the end users. One way to ensure this is to have continuous unit tests forthe product. However, software testing is deemed as an unimportant part of the iteration process by many developers. The extent to which an open-source system hasbeen tested may show the current testing practices or lack thereof. The results couldhelp future developers and product owners to improve the quality of their softwareand help identify bugs early on in the iteration process. Previous studies have shownthat there is a trend amongst developers to avoid testing their products and this research helps to show if that is the case or not. We have created a tool that extractsthe method names, class names, method bodies, test classes and test methods froma project in order to analyse and show how thoroughly they are being tested. Theresults of this project have agreed with the previous studies that there is a tendencyamong developers to avoid testing which can lead to a lot of problems.

Financial reports include a lot of useful information for investors, but extracting this information is time-consuming. We think text summarization is a feasible method. In this thesis, we implement BERTSUM, a state-of-the-art language model for text summarization, and evaluate the results by ROUGE metrics. The experiment was carried out on a novel and large-scale financial dataset called EDGAR-CORPUS. The BERTSUM with a transformer achieves the best performance with a ROUGE-L F1 score of 9.26%. We also hand-picked some model-generated summaries that contained common errors and investigated the causes. The results were then compared to previous research. The ROUGE-L F1 value in the previous study was much higher than ours, we think this is due to the length of the financial reports.

Unsupervised word embedding methods are frequently used for natural language processing applications. However, the unsupervised methods overlook known lexical relations that can be of value to capture accurate semantic word relations. This thesis aims to explore if Swedish word embeddings can benefit from prior known linguistic information. Four knowledge graphs extracted from Svenska Akademiens ordlista (SAOL) are incorporated during the training process using the Probabilistic Word Embeddings with Laplacian Priors (PELP) model. The four implemented PELP models are compared with baseline results to evaluate the use of side information. The results suggest that various lexical relations in SAOL are of interest to generate more accurate Swedish word embeddings.

Historically, cheating in universities has been limited to smuggling notes into exams, unauthorized cooperation, plagiarism and using ghost writers. New improvements in natural language processing now allow students to easily generate text, that is both unique and, in many ways, indistinguishable from what a human would create. These texts can then be submitted with little to no risk of getting caught by anti-cheating software. There are currently a multitude of such text generators online, which vary in ease of use, cost and capabilities. They are capable enough to generate unique text which will evade plagiarism-tools employed by universities. If you combine relatively cheap pricing, ease of use, pressure to perform well in school and low risk of detection. It is not too difficult to imagine that students will use tools like these to cheat. This thesis mainly focuses on whether humans can differentiate AI-generated essays from human written ones and what countermeasures can be used to hinder its use. By giving teachers at Halmstad University human and AI-generated text; then asking them to guess the source of text presented. The experiment concluded that teachers' ability to differentiate AI-generated text from human written text could not be proven.  This thesis also surveys the currently available detection methods for AI-generated text and determines that they are not sufficient in their current form. Lastly, this thesis showcases alternative examination methods that could be used instead of essay-style examinations.