Kernel-Based Ranking. Methods for Learning and Performance Estimation

Machine learning provides tools for automated construction of predictive models in data intensive areas of engineering and science. The family of regularized kernel methods have in the recent years become one of the mainstream approaches to machine learning, due to a number of advantages the methods share. The approach provides theoretically well-founded solutions to the problems of under- and overfitting, allows learning from structured data, and has been empirically demonstrated to yield high predictive performance on a wide range of application domains. Historically, the problems of classification and regression have gained the majority of attention in the field. In this thesis we focus on another type of learning problem, that of learning to rank. In learning to rank, the aim is from a set of past observations to learn a ranking function that can order new objects according to how well they match some underlying criterion of goodness. As an important special case of the setting, we can recover the bipartite ranking problem, corresponding to maximizing the area under the ROC curve (AUC) in binary classification. Ranking applications appear in a large variety of settings, examples encountered in this thesis include document retrieval in web search, recommender systems, information extraction and automated parsing of natural language. We consider the pairwise approach to learning to rank, where ranking models are learned by minimizing the expected probability of ranking any two randomly drawn test examples incorrectly. The development of computationally efficient kernel methods, based on this approach, has in the past proven to be challenging. Moreover, it is not clear what techniques for estimating the predictive performance of learned models are the most reliable in the ranking setting, and how the techniques can be implemented efficiently. The contributions of this thesis are as follows. First, we develop RankRLS, a computationally efficient kernel method for learning to rank, that is based on minimizing a regularized pairwise least-squares loss. In addition to training methods, we introduce a variety of algorithms for tasks such as model selection, multi-output learning, and cross-validation, based on computational shortcuts from matrix algebra. Second, we improve the fastest known training method for the linear version of the RankSVM algorithm, which is one of the most well established methods for learning to rank. Third, we study the combination of the empirical kernel map and reduced set approximation, which allows the large-scale training of kernel machines using linear solvers, and propose computationally efficient solutions to cross-validation when using the approach. Next, we explore the problem of reliable cross-validation when using AUC as a performance criterion, through an extensive simulation study. We demonstrate that the proposed leave-pair-out cross-validation approach leads to more reliable performance estimation than commonly used alternative approaches. Finally, we present a case study on applying machine learning to information extraction from biomedical literature, which combines several of the approaches considered in the thesis. The thesis is divided into two parts. Part I provides the background for the research work and summarizes the most central results, Part II consists of the five original research articles that are the main contribution of this thesis.

Health-related quality of life in school children: validation of instrument, child self assessment, parent-proxy assessment and school nursing documentation of health check-ups

The aims of this study were to validate an international Health-Related Quality of Life (HRQL) instrument, to describe child self and parent-proxy assessed HRQL at child age 10 to 12 and to compare child self assessments with parent-proxy assessments and school nursing documentation. The study is part of the Schools on the Move –research project. In phase one, a cross-cultural translation and validation process was performed to develop a Finnish version of Pediatric Quality of Life Inventory™ 4.0 (PedsQL™ 4.0). The process included a two-way translation, cognitive interviews (children n=7, parents n=5) and a survey (children n=1097, parents n=999). In phase two, baseline and follow-up surveys (children n=986, parents n=710) were conducted to describe and compare the child self and parent-proxy assessed HRQL in school children between the ages 10 and 12. Phase three included two separate data, school nurse documented patient records (children n=270) and a survey (children n=986). The relation between child self assessed HRQL and school nursing documentation was evaluated. Validity and reliability of the Finnish version of PedsQL™ 4.0 was good (Child Self Report α=0.91, Parent-Proxy Report α=0.88). Children reported lower HRQL scores at the emotional (mean 76/80) than the physical (mean 85/89) health domains and significantly lower scores at the age of 10 than 12 (dMean=4, p=<0.001). Agreement between child self and parent-proxy assessment was fragile (r=0,4, p=<0.001) but increased as the child grew from age 10 to 12 years. At health check-ups, school nurses documented frequently children’s physical health, such as growth (97%) and posture (98/99%) but seldom emotional issues, such as mood (2/7%). The PedsQLTM 4.0 is a valid instrument to assess HRQL in Finnish school children although future research is recommended. Children’s emotional wellbeing needs future attention. HRQL scores increase during ages between childhood and adolescence. Concordance between child self and parent-proxy assessed HRQL is low. School nursing documentation, related to child health check-ups, is not in line with child self assessed HRQL and emotional issues need more attention.

Scalable Feature Selection Applications for Genome-Wide Association Studies of Complex Diseases

Personalized medicine will revolutionize our capabilities to combat disease. Working toward this goal, a fundamental task is the deciphering of geneticvariants that are predictive of complex diseases. Modern studies, in the formof genome-wide association studies (GWAS) have afforded researchers with the opportunity to reveal new genotype-phenotype relationships through the extensive scanning of genetic variants. These studies typically contain over half a million genetic features for thousands of individuals. Examining this with methods other than univariate statistics is a challenging task requiring advanced algorithms that are scalable to the genome-wide level. In the future, next-generation sequencing studies (NGS) will contain an even larger number of common and rare variants. Machine learning-based feature selection algorithms have been shown to have the ability to effectively create predictive models for various genotype-phenotype relationships. This work explores the problem of selecting genetic variant subsets that are the most predictive of complex disease phenotypes through various feature selection methodologies, including filter, wrapper and embedded algorithms. The examined machine learning algorithms were demonstrated to not only be effective at predicting the disease phenotypes, but also doing so efficiently through the use of computational shortcuts. While much of the work was able to be run on high-end desktops, some work was further extended so that it could be implemented on parallel computers helping to assure that they will also scale to the NGS data sets. Further, these studies analyzed the relationships between various feature selection methods and demonstrated the need for careful testing when selecting an algorithm. It was shown that there is no universally optimal algorithm for variant selection in GWAS, but rather methodologies need to be selected based on the desired outcome, such as the number of features to be included in the prediction model. It was also demonstrated that without proper model validation, for example using nested cross-validation, the models can result in overly-optimistic prediction accuracies and decreased generalization ability. It is through the implementation and application of machine learning methods that one can extract predictive genotype–phenotype relationships and biological insights from genetic data sets.

New kernel functions and learning methods for text and data mining

Recent advances in machine learning methods enable increasingly the automatic construction of various types of computer assisted methods that have been difficult or laborious to program by human experts. The tasks for which this kind of tools are needed arise in many areas, here especially in the fields of bioinformatics and natural language processing. The machine learning methods may not work satisfactorily if they are not appropriately tailored to the task in question. However, their learning performance can often be improved by taking advantage of deeper insight of the application domain or the learning problem at hand. This thesis considers developing kernel-based learning algorithms incorporating this kind of prior knowledge of the task in question in an advantageous way. Moreover, computationally efficient algorithms for training the learning machines for specific tasks are presented. In the context of kernel-based learning methods, the incorporation of prior knowledge is often done by designing appropriate kernel functions. Another well-known way is to develop cost functions that fit to the task under consideration. For disambiguation tasks in natural language, we develop kernel functions that take account of the positional information and the mutual similarities of words. It is shown that the use of this information significantly improves the disambiguation performance of the learning machine. Further, we design a new cost function that is better suitable for the task of information retrieval and for more general ranking problems than the cost functions designed for regression and classification. We also consider other applications of the kernel-based learning algorithms such as text categorization, and pattern recognition in differential display. We develop computationally efficient algorithms for training the considered learning machines with the proposed kernel functions. We also design a fast cross-validation algorithm for regularized least-squares type of learning algorithm. Further, an efficient version of the regularized least-squares algorithm that can be used together with the new cost function for preference learning and ranking tasks is proposed. In summary, we demonstrate that the incorporation of prior knowledge is possible and beneficial, and novel advanced kernels and cost functions can be used in algorithms efficiently.

Learning Preferences with Kernel-Based Methods

Learning of preference relations has recently received significant attention in machine learning community. It is closely related to the classification and regression analysis and can be reduced to these tasks. However, preference learning involves prediction of ordering of the data points rather than prediction of a single numerical value as in case of regression or a class label as in case of classification. Therefore, studying preference relations within a separate framework facilitates not only better theoretical understanding of the problem, but also motivates development of the efficient algorithms for the task. Preference learning has many applications in domains such as information retrieval, bioinformatics, natural language processing, etc. For example, algorithms that learn to rank are frequently used in search engines for ordering documents retrieved by the query. Preference learning methods have been also applied to collaborative filtering problems for predicting individual customer choices from the vast amount of user generated feedback. In this thesis we propose several algorithms for learning preference relations. These algorithms stem from well founded and robust class of regularized least-squares methods and have many attractive computational properties. In order to improve the performance of our methods, we introduce several non-linear kernel functions. Thus, contribution of this thesis is twofold: kernel functions for structured data that are used to take advantage of various non-vectorial data representations and the preference learning algorithms that are suitable for different tasks, namely efficient learning of preference relations, learning with large amount of training data, and semi-supervised preference learning. Proposed kernel-based algorithms and kernels are applied to the parse ranking task in natural language processing, document ranking in information retrieval, and remote homology detection in bioinformatics domain. Training of kernel-based ranking algorithms can be infeasible when the size of the training set is large. This problem is addressed by proposing a preference learning algorithm whose computation complexity scales linearly with the number of training data points. We also introduce sparse approximation of the algorithm that can be efficiently trained with large amount of data. For situations when small amount of labeled data but a large amount of unlabeled data is available, we propose a co-regularized preference learning algorithm. To conclude, the methods presented in this thesis address not only the problem of the efficient training of the algorithms but also fast regularization parameter selection, multiple output prediction, and cross-validation. Furthermore, proposed algorithms lead to notably better performance in many preference learning tasks considered.

Palkitseminen Palveluyrityksessä

Henkilöstö on yrityksen menestyksen kulmakivi. Henkilöstön suorittaessa oikeita asioita yritys voi saavuttaa tavoitteensa. Suorituskykyperusteisella palkitsemisella taataan oikeudenmukainen ja kannustava palkitseminen. Tämän vuoksi tässä tutkimuksessa tutkitaan suorituskyvyn johtamista ja erilaisia palkitsemismuotoja sekä edellisten johtamista prosessina. Case-tutkimuksena listattuja teoreettisia aihealueita sovelletaan reaalimaailman yrityksessä. Kohdeyrityksessä suoritetaan myös lomakekysely, jolla mahdollistetaan henkilöstön mukaanotto kehitystoimintaan. Palkitseminen on yrityksen keino kannustaa henkilöstöä tehokkuuteen. Palkitsemisen tulee olla mahdollisimman yksinkertaista ja sen on ohjattava yrityksen tavoitteita edesauttaviin suoritteisiin. Palkitsemismuotoja ovat aineeton ja taloudellinen palkitseminen. Palkitsemisen lähtökohtana ovat tavoitteet ja näille asetettavat mittarit, joilla resursseja ohjataan. Esimiehet mahdollistavat tavoitteiden saavuttamisen, jonka vuoksi heitä on koulutettava suorituskyvyn johtamiseen ja palkitsemiseen. Tällöin yrityksen tavoitteet jalkautuvat läpi yrityksen. Osallistamalla henkilöstö palkitsemisen suunnitteluun, palkitsemisesta on saatavissa kannustava ja oikeudenmukainen. Arvioinnissa ja kyselytutkimuksessa kohdeyrityksen suurimmiksi ongelmakohdiksi nousivat viestintä, esimiestoiminta, seuranta, epäoikeudenmukaisuus ja toiminnan kehittymättömyys. Kehitysehdotus kattaa edellisiä ongelmakohtia korjaavia toimenpiteitä, jolloin kohdeyrityksen palkitsemisesta on saatavissa toiminnan ohella kehittyvä, kannustava ja toimintaa ohjaava kokonaisuus.

Reconstruction of cranial bone defects with fiber-reinforced composite–bioactive glass implants

A cranial bone defect may result after an operative treatment of trauma, infection, vascular insult, or tumor. New biomaterials for cranial bone defect reconstructions are needed for example to mimic the biomechanical properties and structure of cranial bone. A novel glass fiber-reinforced composite implant with bioactive glass particulates (FRC–BG, fiber-reinforced composite–bioactive glass) has osteointegrative potential in a preclinical setting. The aim of the first and second study was to investigate the functionality of a FRC–BG implant in the reconstruction of cranial bone defects. During the years 2007–2014, a prospective clinical trial was conducted in two tertiary level academic institutions (Turku University Hospital and Oulu University Hospital) to evaluate the treatment outcome in 35 patients that underwent a FRC–BG cranioplasty. The treatment outcome was good both in adult and pediatric patients. A number of conventional complications related to cranioplasty were observed. In the third study, a retrospective outcome evaluation of 100 cranioplasty procedures performed in Turku University Hospital between years 2002–2012 was conducted. The experimental fourth study was conducted to test the load-bearing capacity and fracture behavior of FRC–BG implants under static loading. The interconnective bars in the implant structure markedly increased the load-bearing capacity of the implant. A loading test did not demonstrate any protrusions of glass fibers or fiber cut. The fracture type was buckling and delamination. In this study, a postoperative complication requiring a reoperation or removal of the cranioplasty material was observed in one out of five cranioplasty patients. The treatment outcomes of cranioplasty performed with different synthetic materials did not show significant difference when compared with autograft. The FRC–BG implant was demonstrated to be safe and biocompatible biomaterial for large cranial bone defect reconstructions in adult and pediatric patients.

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.

Validation of plant dynamic model by online and laboratory measurements – a tool to predict online COD loads out of production of mechanical printing papers

COD discharges out of processes have increased in line with elevating brightness demands for mechanical pulp and papers. The share of lignin-like substances in COD discharges is on average 75%. In this thesis, a plant dynamic model was created and validated as a means to predict COD loading and discharges out of a mill. The assays were carried out in one paper mill integrate producing mechanical printing papers. The objective in the modeling of plant dynamics was to predict day averages of COD load and discharges out of mills. This means that online data, like 1) the level of large storage towers of pulp and white water 2) pulp dosages, 3) production rates and 4) internal white water flows and discharges were used to create transients into the balances of solids and white water, referred to as “plant dynamics”. A conversion coefficient was verified between TOC and COD. The conversion coefficient was used for predicting the flows from TOC to COD to the waste water treatment plant. The COD load was modeled with similar uncertainty as in reference TOC sampling. The water balance of waste water treatment was validated by the reference concentration of COD. The difference of COD predictions against references was within the same deviation of TOC-predictions. The modeled yield losses and retention values of TOC in pulping and bleaching processes and the modeled fixing of colloidal TOC to solids between the pulping plant and the aeration basin in the waste water treatment plant were similar to references presented in literature. The valid water balances of the waste water treatment plant and the reduction model of lignin-like substances produced a valid prediction of COD discharges out of the mill. A 30% increase in the release of lignin-like substances in the form of production problems was observed in pulping and bleaching processes. The same increase was observed in COD discharges out of waste water treatment. In the prediction of annual COD discharge, it was noticed that the reduction of lignin has a wide deviation from year to year and from one mill to another. This made it difficult to compare the parameters of COD discharges validated in plant dynamic simulation with another mill producing mechanical printing papers. However, a trend of moving from unbleached towards high-brightness TMP in COD discharges was valid.

Cross-cultural knowledge transfer within a multinational company

The ability to share knowledge efficiently is vital in day-to-day operations of multinational companies. Therefore, the tools for the effective knowledge sharing along with the transfer knowledge models are of a primary interest for the companies. Knowledge transfer often involves participants of different cultural background. The purpose of this research is to investigate how the cross-cultural knowledge transfer in a multinational corporation takes place and how the successful knowledge transfer is ensured. The four sub-questions will help to answer the research question. Firstly, the author will try to identify what difficulties might occur while transferring knowledge in a multicultural environment. Secondly, the cultural differences’ impact on cross-cultural communication will be explored. Thirdly, the tools and models for effective tacit knowledge transfer will be covered by the author. Lastly, the methods of assessing if a knowledge transfer was successful or not will be discussed. The focus of this thesis is on knowledge transfer within a multinational setting. A multinational corporation is used as a practical example for this study; namely the Training Centre of the multinational corporation. The study was carried out by using a qualitative research approach. The four employees of the Training Centre were interviewed in order to obtain the answers to the research questions. The complete idea of the data gathering process can be obtained by reading through the first part of the methodology section Furthermore, the matters of transferability, confirmability, trustworthiness and objectivity were discussed in the methodological section. The study indicates that in order to carry out the successful and efficient knowledge transfer several requests have to be met. Mutual motivation, e.g. the motivation of both knowledge sender and knowledge receiver is one of the most important determinants of knowledge transfer. Sufficient and bearable workload for the knowledge senders is important to ensure the quality and involvement of the knowledge sender into knowledge transfer process. Course contents, infrastructure and other matters will be discussed further on in relation of successful knowledge transfer. Influences of cultural differences on knowledge sharing were studied in this thesis. At the end of the empirical study- a model used to transfer knowledge in Training Centre will be presented. The model created proves to be functional and appropriate for transferring knowledge to representatives of different cultures.

Strategic management of value networks: how to create value in cross-sector collaboration and partnerships

Cross-sector collaboration and partnerships have become an emerging and desired strategy in addressing huge social and environmental challenges. Despite its popularity, cross-sector collaboration management has proven to be very challenging. Even though cross-sector collaboration and partnership management have been widely studied and discussed in recent years, their effectiveness as well as their ability to create value with respect to the problems they address has remained very challenging. There is little or no evidence of their ability to create value. Regarding all these challenges, this study aims to explore how to manage cross-sector collaborations and partnerships to be able to improve their effectiveness and to create more value for all partners involved in collaboration as well as for customers. The thesis is divided into two parts. The first part comprises an overview of relevant literature (including strategic management, value networks and value creation theories), followed by presenting the results of the whole thesis and the contribution made by the study. The second part consists of six research publications, including both quantitative and qualitative studies. The chosen research strategy is triangulation, as the study includes four types of triangulation: (1) theoretical triangulation, (2) methodological triangulation, (3) data triangulation and (4) researcher triangulation. Two publications represent conceptual development, which are based on secondary data research. One publication is a quantitative study, carried out through a survey. The other three publications represent qualitative studies, based on case studies, where data was collected through interviews and workshops, with participation of managers from all three sectors: public, private and the third (nonprofit). The study consolidates the field of “strategic management of value networks,” which is proposed to be applied in the context of cross-sector collaboration and partnerships, with the aim of increasing their effectiveness and the process of value creation. Furthermore, the study proposes a first definition for the strategic management of value networks. The study also proposes and develops two strategy tools that are recommended to be used for the strategic management of value networks in cross-sector collaboration and partnerships. Taking a step forward, the study implements the strategy tools in practice, aiming to show and to demonstrate how new value can be created by using the developed strategy tools for the strategic management of value networks. This study makes four main contributions. (1) First, it brings a theoretical contribution by providing new insights and consolidating the field of strategic management of value networks, also proposing a first definition for the strategic management of value networks. (2) Second, the study makes a methodical contribution by proposing and developing two strategy tools for value networks of cross-sector collaboration: (a) value network mapping, a method that allows us to assess the current and the potential value network and (b) the Value Network Scorecard, a method of performance measurement and performance prediction in cross-sector collaboration. (3) Third, the study has managerial implications, offering new solutions and empirical evidence on how to increase the effectiveness of cross-sector collaboration and also allow managers to understand how new value can be created in cross-sector partnerships and how to get the full potential of collaboration. (4) And fourth, the study also has practical implications, allowing managers to understand how to use in practice the strategy tools developed in this study, providing discussions on the limitations regarding the proposed tools as well as general limitations involved in the study.

Corporate Entrepreneurship in Healthcare Organizations: An explorative study of employees’ corporate entrepreneurial behavior, job satisfaction and turnover intention

The Finnish healthcare industry is currently facing significant challenges due to economic crises, aging population and major structural reforms, which have resulted in decreased job satisfaction and increased levels of turnover. This proposes that healthcare organizations need to come up with new, creative means to tackle these issues. Several researchers have argued that corporate entrepreneurship may be the necessary means to achieve this. As previous research has mainly focused on examining this concept from organizational perspective, this study looks at how it occurs on the level of individual employees. The purpose of this study is to examine how corporate entrepreneurship is manifested in individual behavior, and how this type of behavior is associated with the individual’s job satisfaction and turnover intention. Additionally, this study will examine the differences in corporate entrepreneurial behavior between private and public sector organizations, as previous research suggests that these two may be characterized differently. Data was collected with the help of a literature review as well as a survey study, which was sent out to a number of employees of four different healthcare organizations, out of which three were public and one was a private sector organization. Six distinct behavioral characteristics were recognized in previous research, which make up the measure for corporate entrepreneurial behavior. Principal components were formed from the different areas of the survey (corporate entrepreneurial behavior, job satisfaction, turnover intention), after which the association of these components were examined with linear regression analysis, which proved that corporate entrepreneurial behavior is positively correlated with both job satisfaction and intention to leave the organization. Differences between sectors were analyzed with analysis of variance and cross tabulation analysis, but neither of these suggested that any significant differences would occur. These results suggest that employees who behave entrepreneurially tend to be more satisfied with their jobs, but also consider leaving their current organizations more often than others. This may be due to the fact that healthcare organizations are not fertile for entrepreneurial behavior, which will drive entrepreneurial individuals looking for employers who may be more supportive of this type of behavior. With growing levels of dissatisfaction as well as little room for entrepreneurial behavior, the studied organizations may actually be in the process of losing those employees who have the ability and desire to behave in such manner, and who could very well be those who will eventually come up with solutions for the major challenges that these organizations are facing.

Yhteistä turvallisuutta rakentaen : poliisi- ja upseeriprofessioiden yhteiskehittelyn mahdollisuuksista

Poliisin rakenneuudistus Pora III ja Puolustusvoimauudistus 2011–2015 ovat olleet viime vuosien esimerkkejä julkisen sektorin työn tehostamisen vaatimuksista. Sekä Puolustusvoimilla että poliisilla, kahdella turvallisuusalan viranomaistaholla, on lakisääteiset tehtävänsä, jotka tulee työn tehostamisen vaatimuksista huolimatta suorittaa. Turvallisuusala yksityistyy vauhdilla, ja keskustelua käydään siitä, mitä tehtäviä viranomainen hoitaa itse, mitä annetaan kaupallisen toimijan tai järjestöjen hoidettavaksi ja mistä kansalainen vastaa itse jatkossa. Resurssi- ja tehostamisvaatimuksia mietittäessä nousee esille upseereiden kohdalla koko maan puolustus ja turvaaminen. Kyetäänkö tämän tehtävän täyttämiseen mahdollisessa sotatilanteessa enää nykyisellä tai mahdollisesti vähenevällä resursoinnilla? Poliisitoimen osalta vasteajat eri puolilla Suomea puhuttavat, samoin se, miten tehtäviä priorisoidaan hoidettavaksi. Saavatko kansalaiset enää perusoikeuksiinsa kuuluvaa arjen turvallisuutta, jonka vielä tänä päivänä katsotaan kuuluvan valtion perustehtäviin? Viranomaisten pitäisi tutkimukseen valittujen aineistojen sekä lakien perusteella hoitaa tehtävänsä laadukkaasti ja tasa-arvoisesti kaikkialla Suomessa. Nykyiset sisäistä ja ulkoista turvallisuutta käsittelevät asiakirjat, esimerkiksi strategiat, puhuvat laajasta turvallisuuskäsityksestä, sisäisen ja ulkoisen turvallisuuden rajojen hämärtymisestä globalisoitumisen seurauksena ja lisääntyvästä poikkihallinnollisesta yhteistyöstä toimintaa ohjaavana ajattelumallina. Viranomaisyhteisyötä tulisi lisätä osana normaalia toimintaa, samoin yhteistyötä järjestöjen, elinkeinoelämän ja jokaisen kansalaisen kanssa. Valtioneuvoston tulevaisuusselonteossa (2013) peräänkuulutetaan uusia, innovatiivisia tapoja hoitaa sekä valtion että kuntien tehtäviä. Yhteisen toiminnan kohteen eli laajan turvallisuuskäsityksen viitekehyksessä on mahdollisuus pohtia uudenlaista turvallisuusalan viranomaisyhteistyötä, eli etsiä perusteluja upseeri- ja poliisiprofession syvemmälle yhteistyölle – yhteiskehittelylle. Tutkimukseni tavoitteena on herätellä keskustelua siitä, onko yhteisen toiminnan kohteen löytymiselle edellytyksiä. Nähtävissä on, että valtiolle kuuluvia toimintoja tehostetaan jatkossakin. Yksi järkevä tapa tehostamisessa on löytää töiden rajapintoja ja yhdistää resurssit näiden osalta. Jotta toiminta olisi tehokasta, sen pitää olla osa jokapäiväistä toimintaa eikä perustua vain muutamiin yhteistoimintaharjoituksiin tai jo tapahtuneiden poikkeustilanteiden hoitoon. Suurin osa kriisiajan toiminnasta perustuu normaaliolojen toimintaan, jolloin sen lähtökohdat voisivat olla yhteisessä työssä ja alkaa jo koulutuksesta, mikä nostetaan tässä työssä yhtenä mahdollisuutena esille. Koulutuksellinen yhteistyö ja liikkuvuus ovat eurooppalaisen tutkintojen viitekehyksen perusteella mahdollista myös kahden eri hallinnonalan koulutuksessa. Tällaista koulutuksen tehostamista haetaan tällä hetkellä muualla yhteiskunnassa. Mutta halutaanko omasta toiminnasta ja tehtävistä luopua edes osittain ja tehdä yhteistyötä mahdollisesti oman työn hallinnan, vallan tai resurssien menettämisen pelossa? Vai onko kyse vain siitä, että yhteistyön syventämiselle ei ole nähty kovinkaan suurta tarvetta tai hyötyä eikä yhteistyö näytä tuovan mitään uutta ammattikuntien osaamiseen? Tutkimuksen aineisto koostuu valtionhallinnon aineistoista, kuten strategioista, mietinnöistä ja raporteista niin sisäisen kuin ulkoisen turvallisuuden alalta. Aineistona käytetään myös upseeri- (n=71) ja poliisipäällystöopiskelijoille (n=65) suunnattua kyselyä ja kirjoitelmaa tulevaisuuden turvallisuusasiantuntijuudesta vuonna 2030. Lisäksi opiskelijavastauksista tehtyä analyysia syvennetään molempien korkeakoulujen (Maanpuolustuskorkeakoulu ja Poliisiammattikorkeakoulu) rehtoreiden sekä molempien hallinnonalojen (puolustusministeriö sekä sisäministeriö) kansliapäälliköiden haastatteluilla. Aineistojen avulla pyritään herättelemään ajatuksia siitä, voisiko yhteisiä töitä löytyä yhteistyön pohjaksi. Tarkoituksena on perustella, miksi yhteistyötä kannattaa tehdä ja ikään kuin vastata etukäteen vastaväitteisiin, miksi sitä ei voitaisi tehdä. Strategioiden yhteistyön tahtotilaa verrataan muihin strategioiden toimenpide ehdotuksiin ja sitä kautta vielä kyselyaineistoon. Opiskelijakyselyllä haetaan näkemyksiä tulevaisuuden turvallisuusasiantuntijuudesta ja mahdollisesta yhteistyöstä sekä sen painopisteistä. Muilla asiantuntijahaastatteluilla haetaan korkeakoulujen sekä ministeriön tason näkemyksiä opiskelijoiden mielipiteisiin. Opiskelijakyselyn avulla on haluttu selvittää sitä, mitä jo työelämässä olleet mutta vaihteeksi opiskelevat sotatieteiden maisteriopiskelijat Maanpuolustuskorkeakoulussa ja poliisin päällystötutkinnon opiskelijat Poliisiammattikorkeakoulussa ajattelevat turvallisuusalan ja -asiantuntijuuden muutoksesta. Minkälaisena he näkevät oman tulevan työnsä ja yhteistyökentän muiden viranomaisten kanssa? Selvää opiskelijavastausten mukaan on se, että turvallisuus halutaan pitää jatkossakin viranomaisen vastuulla ja välttää viimeiseen asti yksityisen sektorin liiallista vastuuta enempää kuin on pakko. Yhteistyötä halutaan edelleen lisätä, ja erityisesti tämä koskee viranomaisten välistä yhteistyötä. Tutkimus on tietoisesti rajattu koskemaan kahta turvallisuusalan viranomaistoimijaa, ammattikorkeakoulutuksen käyneitä poliiseja ja Puolustusvoimien Maanpuolustuskorkeakoulussa opiskelevia upseereita, joiden tehtävistä ja koulutuksesta on löydettävissä yhteisiä rajapintoja ja yhteistyön alueita. Kiinnostus syventyä valittuun kahteen ammattialaan johtuu myös siitä, että usein esimerkiksi sisäasianhallinnon strategioissa Puolustusvoimat jätetään ulkopuolelle varsinkin normaaliolojen yhteistyötä tarkasteltaessa tai vain yksittäisen maininnan asteelle. Sama huomio on havaittavissa puolustushallinnon strategioista. Tämä nousee esille erityisesti alueellista yhteistyötä tai viranomaisyhteistyötä pohdittaessa. Silti sekä sisäministeriön että puolustusministeriön hallinnonalan strategiat ym. perustuvat laajaan turvallisuuskäsitykseen, ja usein eri ammattikuntia analysoitaessa puhutaan tehtävistä, joita tekevät useat ammattikunnat ja professiot. Puhutaan niin sanotuista harmaista alueista. Ministeriöiden tahtotilassa ja toiminnassa on tutkimukseni mukaan nähtävissä ristiriita. Koulutuksen osalta yhteistyön lisääminen on mahdollista etenkin nyt, kun Poliisiammattikorkeakoulussa peruskoulutus on muuttunut ammattikorkeakoulutasoiseksi ja näin tämän ammattikunnan professioasema koulutuksen näkökulmasta on vahvistunut entisestään. Käsittelenkin tutkimuksessani kahta professiota professiotutkimuksen perinteisiä kriteereitä käyttäen, eli rinnastaessani näitä kahta ammattia. Rinnastettavuus koulujen kesken on tullut mahdolliseksi sekä tutkintojen että osaamisen tarkastelun näkökulmasta. Tämän myötä myös molempia korkeakouluja hyödyttävää yhteistyötä olisi mahdollista miettiä osana muutakin hallinnon tehostamista ja rauhan ajan viranomaistoimintaa.