Assessment of an information product project success: case business blueprint

The objective of this study was to find out how project success can be measured in a case where the output of a project is an intangible information product, what kind of framework can be used to evaluate the project success, and how the project assessment can be done in practice. As a case example, the success of a business blueprint project was assessed from the product point of view. A framework for assessing business blueprint project success was made based on a literature review. Furthermore, separate frameworks for measuring information product quality and project costs were developed. The theory of business blueprinting was discovered not to be firmly institutionalized and it is briefly covered in the thesis. The possible net benefits from the strategic business process harmonization were noted to be much more significant than the costs of the business blueprint project. The project was seen as a sufficient success from the viewpoint of the created output.

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.

Self-assessment report for international accreditation - Bachelor's and Master's degree programmes in Information Technology

Lappeenrannan teknillisen yliopiston Tietotekniikan kandidaatin ja maisterin tutkinto-ohjelmien itsearviointi toteutettiin v. 2012 kansainvälistä akkreditointia varten. Itsearviointiraportissa kuvataan tutkinto-ohjelmien tavoitteet, toteutus ja arviointimenettelyt sekä tulokset.

Risk Assessment of a Supplier in the Pharmaceutical Industry

The topic of this Master’s Thesis is risk assessment in the supply chain, and the work was done for a company operating in the pharmaceutical industry. The unique features of the industry bring additional challenges to risk management, due to high regulatory, docu-mentation and traceability requirements. The objective of the thesis was to generate a template for assessing the risks in the supply chain of current and potential suppliers of the case company. Risks pertaining to the case setting were sought mainly from in-house expertise of this specific product and supply chain as well as academic research papers and theory on risk management. A questionnaire was set up to assess the found risks on impact, occurrence and possibility of detection. Through this classification of the severity of the risks, the supplier assessment template was formed. A questionnaire template, comprised of the top 10 risks affecting the flow of information and materials in this setting, was formulated to serve as a generic tool for assessing risks in the supply chain of a pharmaceutical company. The template was tested on another supplier for usability and accuracy of found risks, and it demonstrated functioning in a differing supply chain and product setting.

Parasitic Order Machine. A Sociology and Ontology of Information Securing

This study examines information security as a process (information securing) in terms of what it does, especially beyond its obvious role of protector. It investigates concepts related to ‘ontology of becoming’, and examines what it is that information securing produces. The research is theory driven and draws upon three fields: sociology (especially actor-network theory), philosophy (especially Gilles Deleuze and Félix Guattari’s concept of ‘machine’, ‘territory’ and ‘becoming’, and Michel Serres’s concept of ‘parasite’), and information systems science (the subject of information security). Social engineering (used here in the sense of breaking into systems through non-technical means) and software cracker groups (groups which remove copy protection systems from software) are analysed as examples of breaches of information security. Firstly, the study finds that information securing is always interruptive: every entity (regardless of whether or not it is malicious) that becomes connected to information security is interrupted. Furthermore, every entity changes, becomes different, as it makes a connection with information security (ontology of becoming). Moreover, information security organizes entities into different territories. However, the territories – the insides and outsides of information systems – are ontologically similar; the only difference is in the order of the territories, not in the ontological status of entities that inhabit the territories. In other words, malicious software is ontologically similar to benign software; they both are users in terms of a system. The difference is based on the order of the system and users: who uses the system and what the system is used for. Secondly, the research shows that information security is always external (in the terms of this study it is a ‘parasite’) to the information system that it protects. Information securing creates and maintains order while simultaneously disrupting the existing order of the system that it protects. For example, in terms of software itself, the implementation of a copy protection system is an entirely external addition. In fact, this parasitic addition makes software different. Thus, information security disrupts that which it is supposed to defend from disruption. Finally, it is asserted that, in its interruption, information security is a connector that creates passages; it connects users to systems while also creating its own threats. For example, copy protection systems invite crackers and information security policies entice social engineers to use and exploit information security techniques in a novel manner.

User-generated experiences: how automated platforms for content creation and sharing has transformed information access

Esitys KDK-käytettävyystyöryhmän järjestämässä seminaarissa: Miten käyttäjien toiveet haastavat metatietokäytäntöjämme? / How users' expectations challenge our metadata practices? 30.9.2014.

Linguistic models for decision support

Linguistic modelling is a rather new branch of mathematics that is still undergoing rapid development. It is closely related to fuzzy set theory and fuzzy logic, but knowledge and experience from other fields of mathematics, as well as other fields of science including linguistics and behavioral sciences, is also necessary to build appropriate mathematical models. This topic has received considerable attention as it provides tools for mathematical representation of the most common means of human communication - natural language. Adding a natural language level to mathematical models can provide an interface between the mathematical representation of the modelled system and the user of the model - one that is sufficiently easy to use and understand, but yet conveys all the information necessary to avoid misinterpretations. It is, however, not a trivial task and the link between the linguistic and computational level of such models has to be established and maintained properly during the whole modelling process. In this thesis, we focus on the relationship between the linguistic and the mathematical level of decision support models. We discuss several important issues concerning the mathematical representation of meaning of linguistic expressions, their transformation into the language of mathematics and the retranslation of mathematical outputs back into natural language. In the first part of the thesis, our view of the linguistic modelling for decision support is presented and the main guidelines for building linguistic models for real-life decision support that are the basis of our modeling methodology are outlined. From the theoretical point of view, the issues of representation of meaning of linguistic terms, computations with these representations and the retranslation process back into the linguistic level (linguistic approximation) are studied in this part of the thesis. We focus on the reasonability of operations with the meanings of linguistic terms, the correspondence of the linguistic and mathematical level of the models and on proper presentation of appropriate outputs. We also discuss several issues concerning the ethical aspects of decision support - particularly the loss of meaning due to the transformation of mathematical outputs into natural language and the issue or responsibility for the final decisions. In the second part several case studies of real-life problems are presented. These provide background and necessary context and motivation for the mathematical results and models presented in this part. A linguistic decision support model for disaster management is presented here – formulated as a fuzzy linear programming problem and a heuristic solution to it is proposed. Uncertainty of outputs, expert knowledge concerning disaster response practice and the necessity of obtaining outputs that are easy to interpret (and available in very short time) are reflected in the design of the model. Saaty’s analytic hierarchy process (AHP) is considered in two case studies - first in the context of the evaluation of works of art, where a weak consistency condition is introduced and an adaptation of AHP for large matrices of preference intensities is presented. The second AHP case-study deals with the fuzzified version of AHP and its use for evaluation purposes – particularly the integration of peer-review into the evaluation of R&D outputs is considered. In the context of HR management, we present a fuzzy rule based evaluation model (academic faculty evaluation is considered) constructed to provide outputs that do not require linguistic approximation and are easily transformed into graphical information. This is achieved by designing a specific form of fuzzy inference. Finally the last case study is from the area of humanities - psychological diagnostics is considered and a linguistic fuzzy model for the interpretation of outputs of multidimensional questionnaires is suggested. The issue of the quality of data in mathematical classification models is also studied here. A modification of the receiver operating characteristics (ROC) method is presented to reflect variable quality of data instances in the validation set during classifier performance assessment. Twelve publications on which the author participated are appended as a third part of this thesis. These summarize the mathematical results and provide a closer insight into the issues of the practicalapplications that are considered in the second part of the thesis.

Detecting Deceit – Guessing or Assessing? Study on the Applicability of Veracity Assessment Methods in Human Intelligence

Intelligence from a human source, that is falsely thought to be true, is potentially more harmful than a total lack of it. The veracity assessment of the gathered intelligence is one of the most important phases of the intelligence process. Lie detection and veracity assessment methods have been studied widely but a comprehensive analysis of these methods’ applicability is lacking. There are some problems related to the efficacy of lie detection and veracity assessment. According to a conventional belief an almighty lie detection method, that is almost 100% accurate and suitable for any social encounter, exists. However, scientific studies have shown that this is not the case, and popular approaches are often over simplified. The main research question of this study was: What is the applicability of veracity assessment methods, which are reliable and are based on scientific proof, in terms of the following criteria? o Accuracy, i.e. probability of detecting deception successfully o Ease of Use, i.e. easiness to apply the method correctly o Time Required to apply the method reliably o No Need for Special Equipment o Unobtrusiveness of the method In order to get an answer to the main research question, the following supporting research questions were answered first: What kinds of interviewing and interrogation techniques exist and how could they be used in the intelligence interview context, what kinds of lie detection and veracity assessment methods exist that are reliable and are based on scientific proof and what kind of uncertainty and other limitations are included in these methods? Two major databases, Google Scholar and Science Direct, were used to search and collect existing topic related studies and other papers. After the search phase, the understanding of the existing lie detection and veracity assessment methods was established through a meta-analysis. Multi Criteria Analysis utilizing Analytic Hierarchy Process was conducted to compare scientifically valid lie detection and veracity assessment methods in terms of the assessment criteria. In addition, a field study was arranged to get a firsthand experience of the applicability of different lie detection and veracity assessment methods. The Studied Features of Discourse and the Studied Features of Nonverbal Communication gained the highest ranking in overall applicability. They were assessed to be the easiest and fastest to apply, and to have required temporal and contextual sensitivity. The Plausibility and Inner Logic of the Statement, the Method for Assessing the Credibility of Evidence and the Criteria Based Content Analysis were also found to be useful, but with some limitations. The Discourse Analysis and the Polygraph were assessed to be the least applicable. Results from the field study support these findings. However, it was also discovered that the most applicable methods are not entirely troublefree either. In addition, this study highlighted that three channels of information, Content, Discourse and Nonverbal Communication, can be subjected to veracity assessment methods that are scientifically defensible. There is at least one reliable and applicable veracity assessment method for each of the three channels. All of the methods require disciplined application and a scientific working approach. There are no quick gains if high accuracy and reliability is desired. Since most of the current lie detection studies are concentrated around a scenario, where roughly half of the assessed people are totally truthful and the other half are liars who present a well prepared cover story, it is proposed that in future studies lie detection and veracity assessment methods are tested against partially truthful human sources. This kind of test setup would highlight new challenges and opportunities for the use of existing and widely studied lie detection methods, as well as for the modern ones that are still under development.