Optimization and Applications of Large-Scale Biomedical Event Networks

The overwhelming amount and unprecedented speed of publication in the biomedical domain make it difficult for life science researchers to acquire and maintain a broad view of the field and gather all information that would be relevant for their research. As a response to this problem, the BioNLP (Biomedical Natural Language Processing) community of researches has emerged and strives to assist life science researchers by developing modern natural language processing (NLP), information extraction (IE) and information retrieval (IR) methods that can be applied at large-scale, to scan the whole publicly available biomedical literature and extract and aggregate the information found within, while automatically normalizing the variability of natural language statements. Among different tasks, biomedical event extraction has received much attention within BioNLP community recently. Biomedical event extraction constitutes the identification of biological processes and interactions described in biomedical literature, and their representation as a set of recursive event structures. The 2009–2013 series of BioNLP Shared Tasks on Event Extraction have given raise to a number of event extraction systems, several of which have been applied at a large scale (the full set of PubMed abstracts and PubMed Central Open Access full text articles), leading to creation of massive biomedical event databases, each of which containing millions of events. Sinece top-ranking event extraction systems are based on machine-learning approach and are trained on the narrow-domain, carefully selected Shared Task training data, their performance drops when being faced with the topically highly varied PubMed and PubMed Central documents. Specifically, false-positive predictions by these systems lead to generation of incorrect biomolecular events which are spotted by the end-users. This thesis proposes a novel post-processing approach, utilizing a combination of supervised and unsupervised learning techniques, that can automatically identify and filter out a considerable proportion of incorrect events from large-scale event databases, thus increasing the general credibility of those databases. The second part of this thesis is dedicated to a system we developed for hypothesis generation from large-scale event databases, which is able to discover novel biomolecular interactions among genes/gene-products. We cast the hypothesis generation problem as a supervised network topology prediction, i.e predicting new edges in the network, as well as types and directions for these edges, utilizing a set of features that can be extracted from large biomedical event networks. Routine machine learning evaluation results, as well as manual evaluation results suggest that the problem is indeed learnable. This work won the Best Paper Award in The 5th International Symposium on Languages in Biology and Medicine (LBM 2013).

Mobile marketing process of an international event – Case: Falun2015 FIS Nordic World Ski Championships

Any other technology has never affected daily life at this level and witnessed as speedy adaptation as the mobile phone. At the same time, mobile media has developed to be a serious marketing tool for all kinds of businesses, and the industry has grown explosively in recent years. The objective of this thesis is to inspect the mobile marketing process of an international event. This thesis is a qualitative case study. The chosen case for this thesis is the mobile marketing process of Falun2015 FIS Nordic World Ski Championships due to researcher’s interest on the topic and contacts to the people around the event. The empirical findings were acquired by conducting two interviews with three experts from the case organisation and its partner organisation. The interviews were performed as semi-structured interviews utilising the themes arising from the chosen theoretical framework. The framework distinguished six phases in the process: (i) campaign initiation, (ii) campaign design, (iii) campaign creation, (iv) permission management, (v) delivery, and (vi) evaluation and analysis. Phases one and five were not examined in this thesis because campaign initiation was not purely seen as part of the campaign implementation, and investigating phase five would have required a very technical viewpoint to the study. In addition to the interviews, some pre-established documents were exploited as a supporting data. The empirical findings of this thesis mainly follow the theoretical framework utilised. However, some modifications to the model could be made mainly related to the order of different phases. In the revised model, the actions are categorised depending on the time they should be conducted, i.e. before, during or after the event. Regardless of the categorisation, the phases can be in different order and overlapping. In addition, the business network was highly emphasised by the empirical findings and is thus added to the modified model. Five managerial recommendations can be concluded from the empirical findings of this thesis: (i) the importance of a business network should be highly valued in a mobile marketing process; (ii) clear goals should be defined for mobile marketing actions in order to make sure that everyone involved is aware them; (iii) interactivity should be perceived as part of a mobile marketing communication; (iv) enough time should be allowed for the development of a mobile marketing process in order to exploit all the potential it can offer; and (v) attention should be paid to measuring and analysing matters that are of relevance