Predicting diffusion of innovations with self-organisation and machine learning

The main subject of this master's thesis was predicting diffusion of innovations. The prediction was done in a special case: product has been available in some countries, and based on its diffusion in those countries the prediction is done for other countries. The prediction was based on finding similar countries with Self-Organizing Map~(SOM), using parameters of countries. Parameters included various economical and social key figures. SOM was optimised for different products using two different methods: (a) by adding diffusion information of products to the country parameters, and (b) by weighting the country parameters based on their importance for the diffusion of different products. A novel method using Differential Evolution (DE) was developed to solve the latter, highly non-linear optimisation problem. Results were fairly good. The prediction method seems to be on a solid theoretical foundation. The results based on country data were good. Instead, optimisation for different products did not generally offer clear benefit, but in some cases the improvement was clearly noticeable. The weights found for the parameters of the countries with the developed SOM optimisation method were interesting, and most of them could be explained by properties of the products.

Machine Learning and Clinical Text. Supporting Health Information Flow

Fluent health information flow is critical for clinical decision-making. However, a considerable part of this information is free-form text and inabilities to utilize it create risks to patient safety and cost-­effective hospital administration. Methods for automated processing of clinical text are emerging. The aim in this doctoral dissertation is to study machine learning and clinical text in order to support health information flow.First, by analyzing the content of authentic patient records, the aim is to specify clinical needs in order to guide the development of machine learning applications.The contributions are a model of the ideal information flow,a model of the problems and challenges in reality, and a road map for the technology development. Second, by developing applications for practical cases,the aim is to concretize ways to support health information flow. Altogether five machine learning applications for three practical cases are described: The first two applications are binary classification and regression related to the practical case of topic labeling and relevance ranking.The third and fourth application are supervised and unsupervised multi-class classification for the practical case of topic segmentation and labeling.These four applications are tested with Finnish intensive care patient records.The fifth application is multi-label classification for the practical task of diagnosis coding. It is tested with English radiology reports.The performance of all these applications is promising. Third, the aim is to study how the quality of machine learning applications can be reliably evaluated.The associations between performance evaluation measures and methods are addressed,and a new hold-out method is introduced.This method contributes not only to processing time but also to the evaluation diversity and quality. The main conclusion is that developing machine learning applications for text requires interdisciplinary, international collaboration. Practical cases are very different, and hence the development must begin from genuine user needs and domain expertise. The technological expertise must cover linguistics,machine learning, and information systems. Finally, the methods must be evaluated both statistically and through authentic user-feedback.

Biomedical Event Extraction with Machine Learning

Biomedical natural language processing (BioNLP) is a subfield of natural language processing, an area of computational linguistics concerned with developing programs that work with natural language: written texts and speech. Biomedical relation extraction concerns the detection of semantic relations such as protein-protein interactions (PPI) from scientific texts. The aim is to enhance information retrieval by detecting relations between concepts, not just individual concepts as with a keyword search. In recent years, events have been proposed as a more detailed alternative for simple pairwise PPI relations. Events provide a systematic, structural representation for annotating the content of natural language texts. Events are characterized by annotated trigger words, directed and typed arguments and the ability to nest other events. For example, the sentence “Protein A causes protein B to bind protein C” can be annotated with the nested event structure CAUSE(A, BIND(B, C)). Converted to such formal representations, the information of natural language texts can be used by computational applications. Biomedical event annotations were introduced by the BioInfer and GENIA corpora, and event extraction was popularized by the BioNLP'09 Shared Task on Event Extraction. In this thesis we present a method for automated event extraction, implemented as the Turku Event Extraction System (TEES). A unified graph format is defined for representing event annotations and the problem of extracting complex event structures is decomposed into a number of independent classification tasks. These classification tasks are solved using SVM and RLS classifiers, utilizing rich feature representations built from full dependency parsing. Building on earlier work on pairwise relation extraction and using a generalized graph representation, the resulting TEES system is capable of detecting binary relations as well as complex event structures. We show that this event extraction system has good performance, reaching the first place in the BioNLP'09 Shared Task on Event Extraction. Subsequently, TEES has achieved several first ranks in the BioNLP'11 and BioNLP'13 Shared Tasks, as well as shown competitive performance in the binary relation Drug-Drug Interaction Extraction 2011 and 2013 shared tasks. The Turku Event Extraction System is published as a freely available open-source project, documenting the research in detail as well as making the method available for practical applications. In particular, in this thesis we describe the application of the event extraction method to PubMed-scale text mining, showing how the developed approach not only shows good performance, but is generalizable and applicable to large-scale real-world text mining projects. Finally, we discuss related literature, summarize the contributions of the work and present some thoughts on future directions for biomedical event extraction. This thesis includes and builds on six original research publications. The first of these introduces the analysis of dependency parses that leads to development of TEES. The entries in the three BioNLP Shared Tasks, as well as in the DDIExtraction 2011 task are covered in four publications, and the sixth one demonstrates the application of the system to PubMed-scale text mining.

Evaluation of Machine Learning Classifiers for Mobile Network Intrusion Detection Systems

Mobile malwares are increasing with the growing number of Mobile users. Mobile malwares can perform several operations which lead to cybersecurity threats such as, stealing financial or personal information, installing malicious applications, sending premium SMS, creating backdoors, keylogging and crypto-ransomware attacks. Knowing the fact that there are many illegitimate Applications available on the App stores, most of the mobile users remain careless about the security of their Mobile devices and become the potential victim of these threats. Previous studies have shown that not every antivirus is capable of detecting all the threats; due to the fact that Mobile malwares use advance techniques to avoid detection. A Network-based IDS at the operator side will bring an extra layer of security to the subscribers and can detect many advanced threats by analyzing their traffic patterns. Machine Learning(ML) will provide the ability to these systems to detect unknown threats for which signatures are not yet known. This research is focused on the evaluation of Machine Learning classifiers in Network-based Intrusion detection systems for Mobile Networks. In this study, different techniques of Network-based intrusion detection with their advantages, disadvantages and state of the art in Hybrid solutions are discussed. Finally, a ML based NIDS is proposed which will work as a subsystem, to Network-based IDS deployed by Mobile Operators, that can help in detecting unknown threats and reducing false positives. In this research, several ML classifiers were implemented and evaluated. This study is focused on Android-based malwares, as Android is the most popular OS among users, hence most targeted by cyber criminals. Supervised ML algorithms based classifiers were built using the dataset which contained the labeled instances of relevant features. These features were extracted from the traffic generated by samples of several malware families and benign applications. These classifiers were able to detect malicious traffic patterns with the TPR upto 99.6% during Cross-validation test. Also, several experiments were conducted to detect unknown malware traffic and to detect false positives. These classifiers were able to detect unknown threats with the Accuracy of 97.5%. These classifiers could be integrated with current NIDS', which use signatures, statistical or knowledge-based techniques to detect malicious traffic. Technique to integrate the output from ML classifier with traditional NIDS is discussed and proposed for future work.

Emancipating and developing learning activity: Systemic intervention and re-instrumentation in higher education.

The central theme of this thesis is the emancipation and further development of learning activity in higher education in the context of the ongoing digital transformation of our societies. It was developed in response to the highly problematic mainstream approach to digital re-instrumentation of teaching and studying practises in contemporary higher education. The mainstream approach is largely based on centralisation, standardisation, commoditisation, and commercialisation, while re-producing the general patterns of control, responsibility, and dependence that are characteristic for activity systems of schooling. Whereas much of educational research and development focuses on the optimisation and fine-tuning of schooling, the overall inquiry that is underlying this thesis has been carried out from an explicitly critical position and within a framework of action science. It thus conceptualises learning activity in higher education not only as an object of inquiry but also as an object to engage with and to intervene into from a perspective of intentional change. The knowledge-constituting interest of this type of inquiry can be tentatively described as a combination of heuristic-instrumental (guidelines for contextualised action and intervention), practical-phronetic (deliberation of value-rational aspects of means and ends), and developmental-emancipatory (deliberation of issues of power, self-determination, and growth) aspects. Its goal is the production of orientation knowledge for educational practise. The thesis provides an analysis, argumentation, and normative claim on why the development of learning activity should be turned into an object of individual|collective inquiry and intentional change in higher education, and why the current state of affairs in higher education actually impedes such a development. It argues for a decisive shift of attention to the intentional emancipation and further development of learning activity as an important cultural instrument for human (self-)production within the digital transformation. The thesis also attempts an in-depth exploration of what type of methodological rationale can actually be applied to an object of inquiry (developing learning activity) that is at the same time conceptualised as an object of intentional change within the ongoing digital transformation. The result of this retrospective reflection is the formulation of “optimally incomplete” guidelines for educational R&D practise that shares the practicalphronetic (value related) and developmental-emancipatory (power related) orientations that had been driving the overall inquiry. In addition, the thesis formulates the instrumental-heuristic knowledge claim that the conceptual instruments that were adapted and validated in the context of a series of intervention studies provide means to effectively intervene into existing practise in higher education to support the necessary development of (increasingly emancipated) networked learning activity. It suggests that digital networked instruments (tools and services) generally should be considered and treated as transient elements within critical systemic intervention research in higher education. It further argues for the predominant use of loosely-coupled, digital networked instruments that allow for individual|collective ownership, control, (co-)production, and re-use in other contexts and for other purposes. Since the range of digital instrumentation options is continuously expanding and currently shows no signs of an imminent slow-down or consolidation, individual and collective exploration and experimentation of this realm needs to be systematically incorporated into higher education practise.

Effect of the Knowledge Representation Form on the Efficiency of Knowledge Sharing

The goal of this thesis is to estimate the effect of the form of knowledge representation on the efficiency of knowledge sharing. The objectives include the design of an experimental framework which would allow to establish this effect, data collection, and statistical analysis of the collected data. The study follows the experimental quantitative design. The experimental questionnaire features three sample forms of knowledge: text, mind maps, concept maps. In the interview, these forms are presented to an interviewee, afterwards the knowledge sharing time and knowledge sharing quality are measured. According to the statistical analysis of 76 interviews, text performs worse in both knowledge sharing time and quality compared to visualized forms of knowledge representation. However, mind maps and concept maps do not differ in knowledge sharing time and quality, since this difference is not statistically significant. Since visualized structured forms of knowledge perform better than unstructured text in knowledge sharing, it is advised for companies to foster the usage of these forms in knowledge sharing processes inside the company. Aside of performance in knowledge sharing, the visualized structured forms are preferable due the possibility of their usage in the system of ontological knowledge management within an enterprise.

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.

The role of knowledge management and renewal in building innovation performance

Tämän työn tavoitteena oli selvittää tietojohtamisen eri käytäntöjen vaikutusta oppimiseen, uudistumiseen sekä yrityksen innovaatiokyvykkyyteen. Työssä on keskitytty erityisesti sellaisiin tietojohtamisen käytäntöihin, jotka edistävät oppimista ja uusiutumista yrityksissä. Työssä on käytetty tilastollisia menetelmiä, muun muassa faktorianalyysia, korrelaatioanalyysia sekä regressiota, analysoitaessa 259 suomalaisesta yrityksestä kerättyä kyselydataa niiden tietojohtamisen käytöntöihin ja aineettomaan pääomaan liittyen. Analyysi osoittaa, että useat tietojohtamisen käytännöt vaikuttavat positiivisesti yrityksen uudistumiseen ja sitä kautta innovaatiokyvykkyyteen. Henkilöstön kouluttaminen sekä parhaiden käytäntöjen kerääminen ja soveltaminen yrityksessä ovat positiivisesti yhteydessä innovaatiokyvykkyyteen. Henkilöstön kouluttamisella on merkittävin suora vaikutus innovaatiokyvykkyyteen ja tässä työssä on esitetty, että koulutuksen tarjoamisen suurin vaikutus on oppimismyönteisen kulttuurin kehittyminen yrityksiin sen sijaan, että koulutuksella pyrittäisiin vain parantamaan tehtäväkenttään liittyviä taitoja ja tietoja. Henkilöstön kouluttaminen, parhaat käytännöt sekä sosialisaatiossa tapahtuva tiedon vaihto ja suhteiden solmiminen vaikuttavat positiivisesti uudistumispääomaan. Työn tulosten perusteella uudistumispääomalla on merkittävä rooli innovaatioiden syntymisessä yrityksissä. Uudistumispääoma medioi koulutuksen, parhaiden käytäntöjen ja mahdollisesti myös sosialisaation vaikutusta innovaatiokyvykkyyteen ja on näin merkittävä osa innovaatioiden syntyä yrityksissä. Innovaatiokyvykkyyden osatekijöiden ymmärtäminen voi auttaa johtajia ja esimiehiä keskittämään huomionsa tiettyihin tietojohtamisen käytäntöihin edistääkseen innovaatioiden syntymistä yrityksessä sen sijaan, että he pyrkisivät vain vaikuttamaan innovaatioprosessiin.

Balancing knowledge sharing and protection in collaborative innovation - The roles and dynamics of contracts, IPR's and trust

Pro-gradu tutkielman tavoitteena on tutkia, miten yritykset tasapainoilevat tiedon jakamisen ja suojaamisen välillä innovaatioyhteistyöprojekteissa, ja miten sopimukset, immateriaalioikeudet ja luottamus voivat vaikuttaa tähän tasapainoon. Yhteistyössä yritysten täytyy jakaa tarpeellista tietoa kumppanilleen, mutta toisaalta niiden täytyy varoa, etteivät ne menetä ydinosaamiseensa kuuluvaa tietoa ja kilpailuetuaan. Yrityksillä on useita keinoja tietovuodon estämiseen. Tutkielmassa keskitytään patenttien, sopimusten ja liikesalaisuuksien käyttöön tietoa suojaavina mekanismeina. Kyseiset suojamekanismit vaikuttavat luottamukseen kumppaneiden välillä, ja täten myös näiden halukkuuteen jakaa tietoa kumppaneilleen. Jos kumppanit eivät jaa tarpeeksi tietoa toisilleen, voi yhteistyö epäonnistua. Sopimusten, immateriaalioikeuksien ja luottamuksen rooleja ja vuorovaikutusta tutkitaan kahdenvälisissä yhteistyöprojekteissa. Tutkielmassa esitellään neljä case-esimerkkiä, jotka on koottu suomalaisen metsätoimialan yrityksen haastatteluista.

Supporting knowledge creation and sharing in the early phases of the strategic innovation process

The driving forces of technology and globalization continuously transform the business landscape in a way which undermines the existing strategies and innovations of organizations. The challenge for organizations is to establish such conditions where they are able to create new knowledge for innovative business ideas in interaction between other organizations and individuals. Innovation processes continuously need new external stimulations and seek new ideas, new information and knowledge locating more and more outside traditional organizational boundaries. In several studies, the early phases of the innovation process have been considered as the most critical ones. During these phases, the innovation process can emerge or conclude. External knowledge acquirement and utilization are noticed to be important at this stage of the innovation process giving information about the development of future markets and needs for new innovative businessideas. To make it possible, new methods and approaches to manage proactive knowledge creation and sharing activities are needed. In this study, knowledge creation and sharing in the early phases of the innovation process has been studied, and the understanding of knowledge management in the innovation process in an open and collaborative context advanced. Furthermore, the innovation management methods in this study are combined in a novel way to establish an open innovation process and tested in real-life cases. For these purposes two complementary and sequentially applied group work methods - the heuristic scenario method and the idea generation process - are examined by focusing the research on the support of the open knowledge creation and sharing process. The research objective of this thesis concerns two doctrines: the innovation management including the knowledge management, and the futures research concerning the scenario paradigm. This thesis also applies the group decision support system (GDSS) in the idea generation process to utilize the converged knowledge during the scenario process.

Knowledge creation and transfer in an iterative project environment

This Master’s Thesis examines knowledge creation and transfer processes in an iterative project environment. The aim is to understand how knowledge is created and transferred during an actual iterative implementation project which takes place in International Business Machines (IBM). The second aim is to create and develop new working methods that support more effective knowledge creation and transfer for future iterative implementation projects. The research methodology in this thesis is qualitative. Using focus group interviews as a research method provides qualitative information and introduces the experiences of the individuals participating in the project. This study found that the following factors affect knowledge creation and transfer in an iterative, multinational, and multi-organizational implementation project: shared vision and common goal, trust, open communication, social capital, and network density. All of these received both theoretical and empirical support. As for future projects, strengthening these factors was found to be the key for more effective knowledge creation and transfer.

Role of culture in knowledge transfer and sharing - case ERP implementation

The main objective of this study was to examine how culture influences knowledge transfer and sharing within multicultural ERP project implementation in China. The main interest was to explain how national culture and knowledge are linked by understanding how culture influences knowledge transfer and sharing in a project organization. The other objective of this work was to discuss what Chinese cultural characteristic inhibit and en-hance knowledge sharing in ERP project. The perspective of this study was qualitative and the empirical material was collected from theme interviews among Stora Enso employees. Conclusion of this thesis is that Finns have a very direct style of communication and sharing knowledge whereas Chinese respect face shaving and indirect communication. Another conclusion is that knowledge sharing does not “just happen”, it is needed that project members understand national culture to get all project members commitment to project. In China most important is understand local business processes and understand role of trust and guanxi.

Formal and informal mechanisms for knowledge protection and sharing

The most important knowledge in firms is mostly tacit and embedded in individuals within the organization. This background knowledge that firms possess is used for creation of new knowledge and innovations. As firms today greatly concentrate on their core competencies, they need external knowledge from various collaboration partners. Thus, collaborative relationship governance, as well as control (use of appropriability mechanisms) over background (the input from each firm in innovative activities) and foreground knowledge (the output of collaboration activities) is needed in order to successfully create and capture value from innovative activities without losing core knowledge and competitiveness. Even though research has concentrated on knowledge protection and knowledge sharing, studies that combine both of these views and examine the effects of sharing and protection on value creation and capture have been rather limited. Studies have mainly focused on the protection of the output of innovation while forgetting the protection of the input of innovation. On the other hand, as the research concentrating on the output of innovation tends to favor formal mechanisms, informal mechanisms have remained more unknown to researchers as well as managers. This research aims to combine the perspectives of knowledge sharing and knowledge protection and their relationship with value creation and value capture. The sharing and protection are viewed from two points of view: the use of appropriability mechanisms, as well as governance of the collaborative relationship. The study consists of two parts. The first part introduces the research topic and discusses the overall results. The second part comprises six complementary research publications. Both qualitative and quantitative research methods are used in the study. In terms of results, the findings enhance understanding of the combined use of formal and informal mechanisms for knowledge protection and sharing. Informal mechanisms appear to be emphasized in the protection of background knowledge, and thus are prerequisites for innovation, whereas formal mechanisms are relied on more for protecting the results of innovative activities. However, the simultaneous use of the formal and informal mechanisms that are relevant to the particular industry and innovation context is recommendedthroughout the collaborative innovation process. Further, the study adds to the current knowledge on HRM as an appropriability mechanism: on the firm level its uses include assessing and hedging against employee-related risks such as knowledge leaking and knowledge leaving. A further contribution is to the research on HRM protection and its interrelations with other appropriability mechanisms, its constituents, and its potential use in the area of knowledge protection.