Geo-enabling repositories - techniques, tools, and challenges

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

3D-grafiikan selaintekniikat, erikoistapauksena karttasovellukset

Harava on karttapohjainen kyselypalvelu, jonka avulla voidaan kerätä tietoa erilaisista kyselykohteista. Harava-kyselypalvelussa kyselyihin voidaan vastata muun muassa tekstikentillä, monivalinnoilla ja merkitsemällä alueita ja pisteitä karttapohjaan. Tutkielman tavoitteena oli löytää Harava-kyselypalvelun 2D-karttojen rinnalle 3D-karttavaihtoehto. Aluksi tutkittiin, mitä eri vaihtoehtoja löytyy 3D-komponenttien esittämiseen selaimessa. Tutkituista vaihtoehdoista parhaimmaksi tähän tarkoitukseen osoittautui WebGL-kirjasto. WebGL-kirjaston käyttö suoraan osoittautui vaikeaksi, joten etsittiin rajapintaa, jonka avulla WebGL-kirjaston käyttö helpottuisi. Käsittelyyn otettiin karttapalveluita sekä 3D-mallien esittämiseen tarkoitettuja JavaScript-kirjastoja, jotka käyttävät WebGL-kirjastoa rajapinnan kautta. Näistä sopivimmaksi osoittautui Cesium. Cesium on JavaScript-kirjasto, jonka avulla voidaan toteuttaa 2D-kartta ja 3D-karttapallo sekä upottaa karttapohjaan 3D-elementtejä.

A linkage map of spring turnip rape based on RFLP and RAPD markers

Selostus: RAPD- ja RFLP-markkereista koostuva rypsin kytkentäkartta

Map illustrating the site of the harbours for the proposed Obi-railway : the soundings are all calculated in feet and are based upon the data of the Russian Marine Ministry : the underlined figures are soundings taken by the scooner Bakan, 1887

Kartta kuuluu A. E. Nordenskiöldin kokoelmaan

Matkapuhelinverkon MAP-protokollatoteutus älyverkkopalveluiden ohjauspisteeseen

Työssä esitellään yleiseurooppalaisen GSM-matkapuhelinjärjestelmän verkkoelementtejä ja perehdytään niiden väliseen standardoituun merkinantoprotokollaan. Lisäksi tarkastellaan protokollan lyhytsanomien välitykseen liittyviä operaatioita ja niissä tapahtunutta kehitystä standardoinnin eri vaiheissa. Tavoitteena oli toteuttaa GSM-matkapuhelinverkon merkinantoprotokollaan perustuva ohjelma, jonka tehtävänä on välittää lyhytsanomia matkapuhelin- ja lyhytsanomakeskuksen välillä. Matkapuhelimeen päättyvän lyhytsanoman välitykseen liittyy lisäksi reititystiedon hakeminen vastaanottajan kotirekisteristä. Toteutuksessa on ohjelmointirajapinta, joka helpottaa matkapuhelinverkon uusien palvelusovellusten kehittämistä. Toteutus testattiin standardoituja testitapauksia soveltaen. Yhdenmukaisuustestauksessa käytettiin apuna merkinantoanalysaattoria. Testauksessa tarkastettiin, että protokolla toimii loogisesti oikein. Suorituskykyä ei ole voitu testata todellisessa testiympäristössä, mutta ohjelmallisesti toteutettujen simulaattoreiden avulla on saatu hyviä tuloksia.

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.

Hyperbolic type metrics and distortion of quasiconformal map pings

This Ph.D. thesis consists of four original papers. The papers cover several topics from geometric function theory, more specifically, hyperbolic type metrics, conformal invariants, and the distortion properties of quasiconformal mappings. The first paper deals mostly with the quasihyperbolic metric. The main result gives the optimal bilipschitz constant with respect to the quasihyperbolic metric for the M¨obius self-mappings of the unit ball. A quasiinvariance property, sharp in a local sense, of the quasihyperbolic metric under quasiconformal mappings is also proved. The second paper studies some distortion estimates for the class of quasiconformal self-mappings fixing the boundary values of the unit ball or convex domains. The distortion is measured by the hyperbolic metric or hyperbolic type metrics. The results provide explicit, asymptotically sharp inequalities when the maximal dilatation of quasiconformal mappings tends to 1. These explicit estimates involve special functions which have a crucial role in this study. In the third paper, we investigate the notion of the quasihyperbolic volume and find the growth estimates for the quasihyperbolic volume of balls in a domain in terms of the radius. It turns out that in the case of domains with Ahlfors regular boundaries, the rate of growth depends not merely on the radius but also on the metric structure of the boundary. The topic of the fourth paper is complete elliptic integrals and inequalities. We derive some functional inequalities and elementary estimates for these special functions. As applications, some functional inequalities and the growth of the exterior modulus of a rectangle are studied.

Map matching by using inertial sensors: literature review

This literature review aims to clarify what is known about map matching by using inertial sensors and what are the requirements for map matching, inertial sensors, placement and possible complementary position technology. The target is to develop a wearable location system that can position itself within a complex construction environment automatically with the aid of an accurate building model. The wearable location system should work on a tablet computer which is running an augmented reality (AR) solution and is capable of track and visualize 3D-CAD models in real environment. The wearable location system is needed to support the system in initialization of the accurate camera pose calculation and automatically finding the right location in the 3D-CAD model. One type of sensor which does seem applicable to people tracking is inertial measurement unit (IMU). The IMU sensors in aerospace applications, based on laser based gyroscopes, are big but provide a very accurate position estimation with a limited drift. Small and light units such as those based on Micro-Electro-Mechanical (MEMS) sensors are becoming very popular, but they have a significant bias and therefore suffer from large drifts and require method for calibration like map matching. The system requires very little fixed infrastructure, the monetary cost is proportional to the number of users, rather than to the coverage area as is the case for traditional absolute indoor location systems.