Self-deposit, discovery, and delivery of scientific GIS datasets using GeoHydra

Presentation at Open Repositories 2014, Helsinki, Finland, June 9-13, 2014

Online peer to peer lending: clustering borrowers using self-organizing maps

The purpose of this thesis is to find out whether all the peer to peer lenders are unworthy of credit and also if there are single qualities or combinations of qualities that determine the probability of default of a person or group of people. Distinguishing qualities are searched with self-organizing maps (SOM). Qualities and groups of people found by the self-organizing map are then compared to the average. The comparison is carried out by looking how big proportion of borrowers meeting the criteria is two months or more behind with their payments. Research data used is collected by an Estonian peer to peer lending company during the years of 2011-2014. Data consists of peer to peer borrowers and information gathered from them.

Imbuing airborne electronic warfare self-protection : possibilities to disguise

Tutkielman nimi käännettynä suomeksi on Lentokoneiden omasuojajärjestelmien kyllästäminen – mahdollisuuksia kätkeytymiseen. Siinä keskitytään etsimään aktiivisia, ilmauhkaa kohti säteileviä keinoja ilmahyökkääjän, lentokoneen tai helikopterin, omasuojajärjestelmän harhauttamiseksi. Järjestelmät pyrkivät varoittamaan lentäjää laser- ja tutkasäteilyhavainnoista sekä ohjuksen laukaisusta ja lähestymisestä. Omasuojajärjestelmiin kuuluviin vastakeinoihin tutkimus ei keskity. Tavoitteena harhauttamisella on ilmapuolustuksen todellisen määrän ja sijainnin suojaaminen. Epätietoisuus voi saada lentäjän tekemään vääriä johtopäätöksiä. Työ ei fokusoi harhauttamisen taktiseen kehykseen työn kannalta tarpeellista määrää enempää. Tietosuojasyistä tutkimus ei käsittele käytössä olevien omasuojajärjestelmien operatiivisia ominaisuuksia vaan käsittelee teoriaa niiden taustalla. Tutkimus tehtiin käyttäen taustatutkimukselle enemmän tyypillistä kirjallisuustutkimusta. Jo tutki-muksen alkuvaiheessa oli oletettavaa, että elektronisiin sensoreihin perustuvaa omasuojajärjestelmää pystytään harhauttamaan. Systeemianalysoinnilla pyrittiin löytämään vastauksia tähän olettamukseen. Omasuojajärjestelmästä muodostettiin jo tutkimuksen varhaisessa vaiheessa malli. Tieteellistä kirjallisuutta omasuojajärjestelmistä on olemassa jonkin verran, ja niistä löydettyä tietoa omasuojajärjestelmien ominaisuuksista ja sensoreista yhdistettiin malliin niin, että siitä saatiin mahdollisimman tarkka systeemin kuvaus todellisesta omasuojajärjestelmästä. Analyysin tavoitteena oli löytää niitä kriteereitä, joilla omasuojajärjestelmä saataisiin kohtuullisen tehokkaasti uskomaan harhautusta oikeaksi hälytykseksi. Ohjuksen laukaisusta varoittava sensori perustuu ohjuksen moottorin muodostaman pilven lämpö-säteilyyn. Säteily kuitenkin muuttuu lennon eri vaiheissa, mikä tuottaa haasteita järjestelmälle. Millimetrialueen tutkan käyttö varoittimen sensorina on myös yksi vaihtoehto. Laser-varoittimet toimivat koko sillä taajuusalueella, mitä sotilaskäytössä tulenjohtamiseen ja ohjusten ohjaamiseen käytetään. Tutkavaroittimen tutkimus on vielä kesken. Löydettyjä tuloksia analysoimalla tulen tässä vaiheessa hieman ristiriitaisiin tuloksiin. Lämpösäteilyn käytön suurin haaste on sen eteneminen ilmakehässä. Varoittimen tavoitekaan ei ole toimia kymmentä kilometriä pidemmälle. Yksi mahdollinen ratkaisu on suunnitella ja toteuttaa raketti, jonka tuottaa lämpösäteilyä kuten tietty puolustavan joukon käytössä oleva ohjus. Jos koneessa on kuitenkin myös millimetrialueen tutka tukemassa varoitusjärjestelmää, hankaloituu rakettiharhautus merkittävästi, koska sen pitäisi oletettavasti olla lentokoneen kanssa melko tarkasti kohtaavalla reitillä. Laser-varoitin on ilmeisesti herkin järjestelmistä, koska se voi tietyissä olosuhteissa ja varsinkin matalalla lentokorkeudella aiheuttaa paljon vääriä hälytyksiä ilman tarkoituksellista harhauttamista. Laserin käyttö yhdistettynä raketin laukaisuun saattaisi tuottaa halutun tuloksen. Tutkavaroittimen harhautus onnistuu, jos valelaitteen signaali on uskottavan tarkka.

Digital Humanities For Everyone’s Benefit?: interplay in social media with the great audience, high-esteemed researcher and self-conscious library

The National Library of Finland realizes the Digitization Project of Kindred Languages in 2012–15. The project is financially supported by the Kone Foundation. During this project the National Library of Finland has digitized and made available approximately 1200 monograph and more than 100 newspaper titles in several Uralic languages. The materials are available to both researchers and citizens in the National Library’s Fenno-Ugrica collection. The project will produce digitized materials in the Uralic languages as well as their development tools to support linguistic research and citizen science. The resulting materials will constitute the largest resource for the Uralic languages in the world. Through this project, researchers will gain access to corpora which they have not been able to study before and to which all users will have open access regardless of their place of residence. In my presentation, I will discuss 1) how we utilized the social media (Facebook, Twitter, VKontakte etc) to gain audience for our collection and 2) how the needs of researchers and laymen were met in crowdsourcing.

Designing new learning experiences? : exploring corporate E-learners’ self-regulated learning

The context of this study is corporate e-learning, with an explicit focus on how digital learning design can facilitate self-regulated learning (SRL). The field of e-learning is growing rapidly. An increasing number of corporations use digital technology and elearning for training their work force and customers. E-learning may offer economic benefits, as well as opportunities for interaction and communication that traditional teaching cannot provide. However, the evolving variety of digital learning contexts makes new demands on learners, requiring them to develop strategies to adapt and cope with novel learning tools. This study derives from the need to learn more about learning experiences in digital contexts in order to be able to design these properly for learning. The research question targets how the design of an e-learning course influences participants’ self-regulated learning actions and intentions. SRL involves learners’ ability to exercise agency in their learning. Micro-level SRL processes were targeted by exploring behaviour, cognition, and affect/motivation in relation to the design of the digital context. Two iterations of an e-learning course were tested on two groups of participants (N=17). However, the exploration of SRL extends beyond the educational design research perspective of comparing the effects of the changes to the course designs. The study was conducted in a laboratory with each participant individually. Multiple types of data were collected. However, the results presented in this thesis are based on screen observations (including eye tracking) and video-stimulated recall interviews. These data were integrated in order to achieve a broad perspective on SRL. The most essential change evident in the second course iteration was the addition of feedback during practice and the final test. Without feedback on actions there was an observable difference between those who were instruction-directed and those who were self-directed in manipulating the context and, thus, persisted whenever faced with problems. In the second course iteration, including the feedback, this kind of difference was not found. Feedback provided the tipping point for participants to regulate their learning by identifying their knowledge gaps and to explore the learning context in a targeted manner. Furthermore, the course content was consistently seen from a pragmatic perspective, which influenced the participants’ choice of actions, showing that real life relevance is an important need of corporate learners. This also relates to assessment and the consideration of its purpose in relation to participants’ work situation. The rigidity of the multiple choice questions, focusing on the memorisation of details, influenced the participants to adapt to an approach for surface learning. It also caused frustration in cases where the participants’ epistemic beliefs were incompatible with this kind of assessment style. Triggers of positive and negative emotions could be categorized into four levels: personal factors, instructional design of content, interface design of context, and technical solution. In summary, the key design choices for creating a positive learning experience involve feedback, flexibility, functionality, fun, and freedom. The design of the context impacts regulation of behaviour, cognition, as well as affect and motivation. The learners’ awareness of these areas of regulation in relation to learning in a specific context is their ability for design-based epistemic metareflection. I describe this metareflection as knowing how to manipulate the context behaviourally for maximum learning, being metacognitively aware of one’s learning process, and being aware of how emotions can be regulated to maintain volitional control of the learning situation. Attention needs to be paid to how the design of a digital learning context supports learners’ metareflective development as digital learners. Every digital context has its own affordances and constraints, which influence the possibilities for micro-level SRL processes. Empowering learners in developing their ability for design-based epistemic metareflection is, therefore, essential for building their digital literacy in relation to these affordances and constraints. It was evident that the implementation of e-learning in the workplace is not unproblematic and needs new ways of thinking about learning and how we create learning spaces. Digital contexts bring a new culture of learning that demands attitude change in how we value knowledge, measure it, define who owns it, and who creates it. Based on the results, I argue that digital solutions for corporate learning ought to be built as an integrated system that facilitates socio-cultural connectivism within the corporation. The focus needs to shift from designing static e-learning material to managing networks of social meaning negotiation as part of a holistic corporate learning ecology.

Sentence Compression with Supervised Machine Learning

The subject of the thesis is automatic sentence compression with machine learning, so that the compressed sentences remain both grammatical and retain their essential meaning. There are multiple possible uses for the compression of natural language sentences. In this thesis the focus is generation of television program subtitles, which often are compressed version of the original script of the program. The main part of the thesis consists of machine learning experiments for automatic sentence compression using different approaches to the problem. The machine learning methods used for this work are linear-chain conditional random fields and support vector machines. Also we take a look which automatic text analysis methods provide useful features for the task. The data used for machine learning is supplied by Lingsoft Inc. and consists of subtitles in both compressed an uncompressed form. The models are compared to a baseline system and comparisons are made both automatically and also using human evaluation, because of the potentially subjective nature of the output. The best result is achieved using a CRF - sequence classification using a rich feature set. All text analysis methods help classification and most useful method is morphological analysis. Tutkielman aihe on suomenkielisten lauseiden automaattinen tiivistäminen koneellisesti, niin että lyhennetyt lauseet säilyttävät olennaisen informaationsa ja pysyvät kieliopillisina. Luonnollisen kielen lauseiden tiivistämiselle on monta käyttötarkoitusta, mutta tässä tutkielmassa aihetta lähestytään television ohjelmien tekstittämisen kautta, johon käytännössä kuuluu alkuperäisen tekstin lyhentäminen televisioruudulle paremmin sopivaksi. Tutkielmassa kokeillaan erilaisia koneoppimismenetelmiä tekstin automaatiseen lyhentämiseen ja tarkastellaan miten hyvin erilaiset luonnollisen kielen analyysimenetelmät tuottavat informaatiota, joka auttaa näitä menetelmiä lyhentämään lauseita. Lisäksi tarkastellaan minkälainen lähestymistapa tuottaa parhaan lopputuloksen. Käytetyt koneoppimismenetelmät ovat tukivektorikone ja lineaarisen sekvenssin mallinen CRF. Koneoppimisen tukena käytetään tekstityksiä niiden eri käsittelyvaiheissa, jotka on saatu Lingsoft OY:ltä. Luotuja malleja vertaillaan Lopulta mallien lopputuloksia evaluoidaan automaattisesti ja koska teksti lopputuksena on jossain määrin subjektiivinen myös ihmisarviointiin perustuen. Vertailukohtana toimii kirjallisuudesta poimittu menetelmä. Tutkielman tuloksena paras lopputulos saadaan aikaan käyttäen CRF sekvenssi-luokittelijaa laajalla piirrejoukolla. Kaikki kokeillut teksin analyysimenetelmät auttavat luokittelussa, joista tärkeimmän panoksen antaa morfologinen analyysi.

Visual Customer Segmentation and Data Exploration with the Self-Organizing Map

Advancements in information technology have made it possible for organizations to gather and store vast amounts of data of their customers. Information stored in databases can be highly valuable for organizations. However, analyzing large databases has proven to be difficult in practice. For companies in the retail industry, customer intelligence can be used to identify profitable customers, their characteristics, and behavior. By clustering customers into homogeneous groups, companies can more effectively manage their customer base and target profitable customer segments. This thesis will study the use of the self-organizing map (SOM) as a method for analyzing large customer datasets, clustering customers, and discovering information about customer behavior. Aim of the thesis is to find out whether the SOM could be a practical tool for retail companies to analyze their customer data.

Business intelligence – enabling self-service: A case study in a global manufacturing company

Business intelligence (BI) is an information process that includes the activities and applications used to transform business data into valuable business information. Today’s enterprises are collecting detailed data which has increased the available business data drastically. In order to meet changing customer needs and gain competitive advantage businesses try to leverage this information. However, IT departments are struggling to meet the increased amount of reporting needs. Therefore, recent shift in the BI market has been towards empowering business users with self-service BI capabilities. The purpose of this study was to understand how self-service BI could help businesses to meet increased reporting demands. The research problem was approached with an empirical single case study. Qualitative data was gathered with a semi-structured, theme-based interview. The study found out that case company’s BI system was mostly used for group performance reporting. Ad-hoc and business user-driven information needs were mostly fulfilled with self-made tools and manual work. It was felt that necessary business information was not easily available. The concept of self-service BI was perceived to be helpful to meet such reporting needs. However, it was found out that the available data is often too complex for an average user to fully understand. The respondents felt that in order to self-service BI to work, the data has to be simplified and described in a way that it can be understood by the average business user. The results of the study suggest that BI programs struggle in meeting all the information needs of today’s businesses. The concept of self-service BI tries to resolve this problem by allowing users easy self-service access to necessary business information. However, business data is often complex and hard to understand. Self-serviced BI has to overcome this challenge before it can reach its potential benefits.

Business intelligence – enabling self-service: A case study in a global manufacturing company

Business intelligence (BI) is an information process that includes the activities and applications used to transform business data into valuable business information. Today’s enterprises are collecting detailed data which has increased the available business data drastically. In order to meet changing customer needs and gain competitive advantage businesses try to leverage this information. However, IT departments are struggling to meet the increased amount of reporting needs. Therefore, recent shift in the BI market has been towards empowering business users with self-service BI capabilities. The purpose of this study was to understand how self-service BI could help businesses to meet increased reporting demands. The research problem was approached with an empirical single case study. Qualitative data was gathered with a semi-structured, theme-based interview. The study found out that case company’s BI system was mostly used for group performance reporting. Ad-hoc and business user-driven information needs were mostly fulfilled with self-made tools and manual work. It was felt that necessary business information was not easily available. The concept of self-service BI was perceived to be helpful to meet such reporting needs. However, it was found out that the available data is often too complex for an average user to fully understand. The respondents felt that in order to self-service BI to work, the data has to be simplified and described in a way that it can be understood by the average business user. The results of the study suggest that BI programs struggle in meeting all the information needs of today’s businesses. The concept of self-service BI tries to resolve this problem by allowing users easy self-service access to necessary business information. However, business data is often complex and hard to understand. Self-serviced BI has to overcome this challenge before it can reach its potential benefits.

Expressing the self in postmodern travel consumption: a case study of students

The study approaches student travel from the perspective of postmodern consumption. The background is in the observation that the student travel market has a vast potential, but it is not necessarily capitalized upon to the extent it could. This might partly have to do with the peculiarities of postmodernity: consumption is characterized by unpredictability and abstract motives. The research questions are built around what constitutes student travel consumption and how can students be categorized according to motivation, behaviour and values. Also identity and expressiveness are present and it is evaluated, if travel services facilitate these background is the observation that the student travel market has a vast potential, but it is not necessarily capitalized upon to the extent it could be. This might partly have to do with the peculiarities of postmodernity: consumption is characterized by unpredictability and abstract motives. The research questions are built around what constitutes student travel consumption and how can students be categorized according to motivation, behaviour and values. Also identity and expressiveness are present and it is evaluated whether travel services facilitate these constructs. The topic is approached by discovering the key concepts such as self-identity. This was done in order to create survey questions that reflect the underlying theories. The survey was sent to chosen student groups of Turku School of Economics. The data was analyzed using statistical methods, mainly principal component analysis, in order to categorize students’ motives and behaviour into distinct profiles. The findings indicate that students have a high level of awareness in their travel consumption choices. Travel services seem to facilitate identity and lifestyle expressiveness, one central dimension of postmodernity. Psychographics such as motivation seem to work well as a segmentation criteria when it comes to the student traveler market. Travel offers students an opportunity for relaxation, escape, enjoyment and gaining new experiences and social contacts. Furthermore, the enjoyment of the travel experience extends to the pre- and post-trip time.