Internet-lähteistä haettujen uutisten ryhmittely

Tämän diplomityön tarkoituksena on tutkia, mitä vaaditaan uutisten samanlaisuuden automaattiseen tunnistamiseen. Uutiset ovat tekstipohjaisia uutisia, jotka on haettu eri uutislähteistä. Uutisista on tarkoitus tunnistaa ensinnäkin ne uutiset, jotka tarkoittavat samaa asiaa, sekä ne uutiset, jotka eivät ole aivan sama asia, mutta liittyvät kuitenkin toisiinsa. Tässä diplomityössä tutkitaan, millä algoritmeilla tämä tunnistus onnistuu tehokkaimmin sekä suomalaisessa, että englanninkielisessä tekstissä. Diplomityössä vertaillaan valmiita algoritmeja. Tavoitteena on valita sellainen algoritmiyhdistelmä, että 90 % vertailluista uutisista tunnistuu oikein. Tutkimuksessa käytetään 2 eri ryhmittelyalgoritmia, sekä 3 eri stemmaus-algoritmia. Näitä algoritmeja vertaillaan sekä uutisten tunnistustehokkuuden, että niiden suorituskyvyn suhteen. Parhaimmaksi stemmaus-algoritmiksi osoittautui sekä suomen-, että englanninkielisten uutisten vertailussa Porterin algoritmi. Ryhmittely-algoritmeista tehokkaammaksi osoittautui yksinkertaisempi erilaisiin tunnuslukuihin perustuva algoritmi.

Automatic system identification and optimization of centrifugal pump operation

Centrifugal pumps are a notable end-consumer of electrical energy. Typical application of a centrifugal pump is the filling or emptying of a reservoir tank, where the pump is often operated at a constant speed until the process is completed. Installing a frequency converter to control the motor substitutes the traditional fixed-speed pumping system, allows the optimization of rotational speed profile for the pumping tasks and enables the estimation of rotational speed and shaft torque of an induction motor without any additional measurements from the motor shaft. Utilization of variable-speed operation provides the possibility to decrease the overall energy consumption of the pumping task. The static head of the pumping process may change during the pumping task. In such systems, the minimum rotational speed changes during reservoir filling or emptying, and the minimum energy consumption can’t be achieved with a fixed rotational speed. This thesis presents embedded algorithms to automatically identify, optimize and monitor pumping processes between supply and destination reservoirs, and evaluates the changing static head –based optimization method.

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).

Use of pyrosequencing to identify streptococci and to detect mutations causing antimicrobial resistance

Rapid identification and resistance determination of pathogens in clinical specimens is vital for accurate treatment and monitoring of infectious diseases. Antimicrobial drug resistance is increasing globally and healthcare settings are facing this cost-intensive and even life-threatening problem. The incidence of resistant pathogens in Finland has remained relatively steady and manageable at least for the time being. DNA sequencing is the gold standard method for genotyping, mutation analysis, and identification of bacteria. Due to significant cost decrease in recent years, this technique is available to many research and clinical laboratories. Pyrosequencing technique, a rapid real-time DNA sequencing method especially suitable for analyzing fairly short stretches of DNA, was used in this study. Due to its robustness and versatility, pyrosequencing was applied in this study for identification of streptococci and detection of certain mutations causing antimicrobial resistance in different bacteria. Certain streptococcal species such as S. pneumoniae and S. pyogenes are significantly important clinical pathogens. S. pneumoniae causes e.g. pneumonia and otitis media and is one of the most important community-acquired pathogens. S. pyogenes, also known as group A streptococcus, causes e.g. angina and erysipelas. In contrast, the socalled alpha-haemolytic streptococci, such as S. mitis and S. oralis, belong to the normal microbiota, which are regarded to be non-pathogenic and are nearly impossible to identify by phenotypic methods. In this thesis, a pyrosequencing method was developed for identification of streptococcal species based on the 16S rRNA sequences. Almost all streptococcal species could be differentiated from one another by the developed method, including S. pneumoniae from its close relatives S. mitis and S. oralis . New resistance genes and their variants are constantly discovered and reported. In this study, new methods for detecting certain mutations causing macrolide resistance or extended spectrum beta-lactamase (ESBL) phenotype were developed. These resistance detection approaches are not only suitable for surveillance of mechanisms causing antimicrobial resistance but also for routine analysis of clinical samples particularly in epidemic settings. In conclusion, pyrosequencing was found to be an accurate, versatile, cost-effective, and rapid DNA sequencing method that is especially suitable for mutation analysis of short DNA fragments and identification of certain bacteria.

Julkisten tukitoimien vaikutus sähköautojen yleistymiseen

Tässä kandidaatintyössä tutkittiin julkisten toimijoiden asettamia sähköautojen yleistymistavoitteita kymmenessä maassa. Tutkimus toteutettiin kirjallisuusselvityksenä. Tavoitteena oli selvittää sähköautojen yleistymiseen vaikuttavia tekijöitä. Erityisenä kiinnostuksen kohteena olivat julkishallintojen sähköauton hankintaan kohdistamat kannustimet eri maissa ja se, onko kannustimien voimakkuudella havaittavissa olevaa vaikutusta sähköautojen yleistymiseen. Tutkimuksessa luotiin myös katsaus sähköautojen myyntiin ja yleistymisodotuksiin Suomessa ja verrattiin niitä muihin maihin. Tutkimuksessa havaittiin, että kannustimien voimakkuudella ei ole selvää yhteyttä sähköautojen yleistymiseen. On mahdollista, että kannustimet vaikuttavat positiivisesti sähköautojen myyntiin. Vaikuttavia tekijöitä on kuitenkin muitakin, ja niiden vaikutus voi olla suurempi kuin julkisten kannustimien. Suomessa sähköautojen myynti on samassa suuruusluokassa kuin Keski-Euroopassa, mutta merkittävästi jäljessä muita Pohjoismaita. Myös kannustimien voimakkuudella mitattuna Suomi häviää muille Pohjoismaille ja lisäksi useille Euroopan maille. Voimakkaammat tukimuodot voisivat kasvattaa sähköautojen myyntiä Suomessa, mutta julkisen talouden alijäämäisyys ei luultavasti mahdollista nykyistä voimakkaampien tukien käyttöönottoa.

Communicating trustworthiness in online stores

Even though e-commerce systems are expected to have many advantages compared to the traditional ways of doing business, it is not always the reality. Lack of trust is still said to be one of the most important barriers to online shopping. In traditional stores, trust has usually been established in a direct contact between the customer and the company or its personnel. In online stores, there is no direct interaction. The purpose of this thesis is to identify the key antecedents to online trust and to distinguish between effective and ineffective practices. A model on how consumers establish initial trust towards an unknown online vendor was proposed based on previous theories. The model was tested empirically by targeting an online survey at higher degree students in Finland and in Germany. The data confirmed the proposed view that trusting intentions are affected by individual characteristics, characteristics of the company as well as characteristics of the website. Additionally national differences were found between Finnish and German respondents. The data suggested that online vendors can convey a message of trustworthiness by improving information quality and overall usefulness of the website. Perceived risk of online shopping was found to depend especially on general trust in the Internet, service quality and ease of use. A trustworthy online store should include several payment methods as well as means to access and modify given data. The vendors should also make sure that inquiries are addressed quickly, transactions are confirmed automatically and that customers have a possibility to track their order. A model that includes three different sources of trust should contribute to the theoretical understanding of trust formation in online stores. The resulting list of trust antecedents can also be used as a checklist when e-commerce practitioners wish to optimize the trust building.

Data Governance and Automated Marketing – A Case Study of Expected Benefits of Organizing Data Governance in an ICT Company

This research is looking to find out what benefits employees expect the organization of data governance gains for an organization and how it benefits implementing automated marketing capabilities. Quality and usability of the data are crucial for organizations to meet various business needs. Organizations have more data and technology available what can be utilized for example in automated marketing. Data governance addresses the organization of decision rights and accountabilities for the management of an organization’s data assets. With automated marketing it is meant sending a right message, to a right person, at a right time, automatically. The research is a single case study conducted in Finnish ICT-company. The case company was starting to organize data governance and implementing automated marketing capabilities at the time of the research. Empirical material is interviews of the employees of the case company. Content analysis is used to interpret the interviews in order to find the answers to the research questions. Theoretical framework of the research is derived from the morphology of data governance. Findings of the research indicate that the employees expect the organization of data governance among others to improve customer experience, to improve sales, to provide abilities to identify individual customer’s life-situation, ensure that the handling of the data is according to the regulations and improve operational efficiency. The organization of data governance is expected to solve problems in customer data quality that are currently hindering implementation of automated marketing capabilities.