Semantic Classes in Topic Detection and Tracking

Topic detection and tracking (TDT) is an area of information retrieval research the focus of which revolves around news events. The problems TDT deals with relate to segmenting news text into cohesive stories, detecting something new, previously unreported, tracking the development of a previously reported event, and grouping together news that discuss the same event. The performance of the traditional information retrieval techniques based on full-text similarity has remained inadequate for online production systems. It has been difficult to make the distinction between same and similar events. In this work, we explore ways of representing and comparing news documents in order to detect new events and track their development. First, however, we put forward a conceptual analysis of the notions of topic and event. The purpose is to clarify the terminology and align it with the process of news-making and the tradition of story-telling. Second, we present a framework for document similarity that is based on semantic classes, i.e., groups of words with similar meaning. We adopt people, organizations, and locations as semantic classes in addition to general terms. As each semantic class can be assigned its own similarity measure, document similarity can make use of ontologies, e.g., geographical taxonomies. The documents are compared class-wise, and the outcome is a weighted combination of class-wise similarities. Third, we incorporate temporal information into document similarity. We formalize the natural language temporal expressions occurring in the text, and use them to anchor the rest of the terms onto the time-line. Upon comparing documents for event-based similarity, we look not only at matching terms, but also how near their anchors are on the time-line. Fourth, we experiment with an adaptive variant of the semantic class similarity system. The news reflect changes in the real world, and in order to keep up, the system has to change its behavior based on the contents of the news stream. We put forward two strategies for rebuilding the topic representations and report experiment results. We run experiments with three annotated TDT corpora. The use of semantic classes increased the effectiveness of topic tracking by 10-30\% depending on the experimental setup. The gain in spotting new events remained lower, around 3-4\%. The anchoring the text to a time-line based on the temporal expressions gave a further 10\% increase the effectiveness of topic tracking. The gains in detecting new events, again, remained smaller. The adaptive systems did not improve the tracking results.

Ihanteena vapaaehtoisuus : Diakoniatyö Porvoon hiippakunnan seurakunnissa, erityisesti Iitin ja Tampereen rovastikunnissa vuosina 1897-1923

The Ideal of Volunteerism. An institutional approach to social welfare work in the parishes of the Diocese of Porvoo especially in the deaneries of Iitti and Tampere, Finland, in the years 1897-1923 Social welfare work (also known as diakonia) has achieved a high status in the Evangelical Lutheran Church of Finland. Since 1944, provisions of the Finnish Church Act have obliged each parish to employ at least one deacon or deaconess. This study sets out to examine the background and development of social welfare work in the Evangelical Lutheran Church of Finland from the 1890s to the 1920s, by which time social welfare work had become an established practice in the Church. The study investigates the development of social welfare work on the level of parishes. The main source material was collected from sixteen parishes in the Diocese of Porvoo especially in the deaneries of Iitti and Tampere. In the 1890s, two approaches were used in church social work in Finland. The dioceses of Kuopio, Savonlinna and Turku pursued a congregational approach to social work, while the Diocese of Porvoo employed an institutional approach, mainly because of the influence of Bishop Herman Råbergh. This study charts the formation of church social work in Finnish parishes, which took place during a period of tension between the two approaches. The institutional approach to church social work adopted by the Diocese of Porvoo was based on the German system of Asisters= houses@, in which deaconess institutes sent parish sisters to serve congregations. The parish or, in many cases, a separate association dedicated to church social work paid an annual fee to the deaconess institute, which took care of the parish sisters in old age. In the institutional approach, volunteers were recruited to carry out church social work. It was considered as inappropriate to use tax revenue or other public funding for church social work, which was supposed to be based on Christian love for one=s fellow humans and the needy, and for which only voluntary financial contributions were supposed to be used. In the congregational approach, church social work was directly based on the efforts of the parish. The approach relied on the administrative bodies of parishes and the Church, and tax revenue collected by the parishes, as well as other forms of public funding, could be used to carry out the social welfare work. The parishes employed deacons and deaconesses and paid their salaries. The approaches described above were not pursued in their ideal forms; instead, many variations existed. However, in principle, the social welfare work undertaken by the parishes of the Diocese of Porvoo was based on the institutional approach, while the congregational approach was largely employed elsewhere in Finland. Both of the approaches were viable. Parishes began to employ deacons and deaconesses as of the 1890s. The number of parishes which had hired a deacon or deaconess increased particularly in the 1910s, by which time 60% of parishes had employed one. This level was maintained until 1944 when each parish in the Evangelical Lutheran Church of Finland was obliged to employ a deacon or deaconess. Deaconesses usually worked as travelling nurses. The autonomous status of Finland as part of the Russian Empire did not give Finns the right to develop legislation on social affairs and health care. Consequently, the legislation process did not begin until Finland gained its independence in 1917. The social welfare work carried out by parishes and a number of voluntary organisations satisfied the emerging need for medical treatment in Finnish society. Neither the government nor the municipalities had sufficient resources to provide this treatment. Based on the ideal of volunteerism, the institutional social work practiced in the Diocese of Porvoo ran into serious difficulties at the end of the First World War. Because of severe inflation, prices began to rise as of 1915 and tripled in 1917-1918. During the same period, Finnish society went through a deep crisis which escalated into Civil War in spring 1918. This period of economic and social turmoil marked a turning-point which led to a weakening of the status of institutional social work in parishes. Voluntary efforts were no longer sufficient to maintain the practice. In contrast, congregational social work, which was based on public funding, was able to cope with the changes and survived the crisis. The approach to social work adopted by the Diocese of Porvoo turned out to be no more than a brief detour in the history of social work in the Evangelical Lutheran Church of Finland. At the start of the 1920s, the two approaches were integrated into a common vision for establishing church social work as a statutory practice in parishes.

Short- and long-term consequences of food resources on Ural owl Strix uralensis reproduction

Life-history theory states that although natural selection would favour a maximisation of both reproductive output and life-span, such a combination can not be achieved in any living organism. According to life-history theory the reason for the fact that not all traits can be maximised simultaneously is that different traits compete with each other for resources. These relationships between traits that constrain the simultaneous evolution of two or more traits are called trade-offs. Therefore, during different life-stages an individual needs to optimise its allocation of resources to life-history components such as growth, reproduction and survival. Resource limitation acts on these traits and therefore investment in one trait, e.g. reproduction, reduces the resources available for investment in another trait, e.g. residual reproduction or survival. In this thesis I study how food resources during different stages of the breeding event affect reproductive decisions in the Ural owl (Strix uralensis) and the consequences of these decisions on parents and offspring. The Ural owl is a suitable study species for such studies in natural populations since they are long-lived, site-tenacious, and feed on voles. The vole populations in Fennoscandia fluctuate in three- to four-year cycles, which create a variable food environment for the Ural owls to cope with. The thesis gives new insight in reproductive costs and their consequences in natural animal populations with emphasis on underlying physiological mechanisms. I found that supplementary fed Ural owl parents invest supplemented food resources during breeding in own self-maintenance instead of allocating those resources to offspring growth. This investment in own maintenance instead of improving current reproduction had carry-over effects to the following year in terms of increased reproductive output. Therefore, I found evidence that reduced reproductive costs improves future reproductive performance. Furthermore, I found evidence for the underlying mechanism behind this carry-over effect of supplementary food on fecundity. The supplementary-fed parents reduced their feeding investment in the offspring compared to controls, which enabled the fed female parents to invest the surplus resources in parasite resistance. Fed female parents had lower blood parasite loads than control females and this effect lasted until the following year when also reproductive output was increased. Hence, increased investment in parasite resistance when resources are plentiful has the potential to mediate positive carry-over effects on future reproduction. I further found that this carry-over effect was only present when potentials for future reproduction were good. The thesis also provides new knowledge on resource limitation on maternal effects. I found that increased resources prior to egg laying improve the condition and health of Ural owl females and enable them to allocate more resources to reproduction than control females. These additional resources are not allocated to increase the number of offspring, but instead to improve the quality of each offspring. Fed Ural owl females increased the size of their eggs and allocated more health improving immunological components into the eggs. Furthermore, the increased egg size had long-lasting effects on offspring growth, as offspring from larger eggs were heavier at fledging. Limiting resources can have different short- and long-term consequences on reproductive decisions that affect both offspring number and quality. In long-lived organisms, such as the Ural owl, it appears to be beneficial in terms of fitness to invest in long breeding life-span instead of additional investment in current reproduction. In Ural owls, females can influence the phenotypic quality of the offspring by transferring additional resources to the eggs that can have long-lasting effects on growth.

The Rocky Road of Social Sustainability : The Impact of Integrated Biodiversity Conservation and Development on the Local Realities in Ranomafana National Park, Madagascar

The objective of this study is to examine the social impacts of the integrated conservation and development project (ICDP) aimed at biodiversity conservation and local socio-economic development in the Ranomafana National Park (RNP), Madagascar. Furthermore, the study explores social sustainability and justice of the ICDP in Ranomafana. This ethnographically informed impact study uses of various field methods. The research material used consists of observation, interviews (key-person and focus group), school children's writings, official statistics and project documents. Fieldwork was conducted in three phases in 2001, 2002 and 2004 in twelve villages around the park, as well as in neighbouring areas of Ranomafana. However, four of those twelve villages were chosen for closer study. This study consists of five independent articles and a concluding chapter. Social impacts were studied through reproductive health indicators as well as a life security approach. Equity and distribution of benefits and drawbacks of ICDP were analysed and the actors related to the conservation in Ranomafana were identified. The children and adolescents' environmental views were also examined. The reproductive health indicators studied showed a poor state of reproductive health in the park area. Moreover, the existing social capital in the villages seemed to be fragmented due to economic difficulties that were partly caused by the conservation regulations. The ICDP in Ranomafana did not pay attention to the heterogeneity of the affected communities even though the local beneficiaries of the ICDP varied according to their ethnicity, living place, wealth, social position and gender. In addition, various conservation actors (local people in various groups, local authorities, tourist business owners, conservation NGOs and scientists) contest their interests over the forest, conservation and its related activities. This study corroborates the same type of evidence and conclusions discussed in other similar cases elsewhere: so called social conservation programmes still cannot meet the needs of the people living near the protected areas; on the contrary, they even have a reverse impact on the people's lives. A fundamental misunderstood assumption in the conservation process in Ranomafana was to consider the local people as a problem for biodiversity conservation. Major reasons for the failure of the ICDP in Ranomafana include a lack of local institutions that would have been able to communicate as equals with the conservation NGOs as well as to transfer the tradition of the authoritarian governance in conservation management together with the over-appreciation of scientific biodiversity, and lack of will to understand the local people's rights to use the forest for their livelihoods.

Food biotechnologies in Italy: a social psychological study

This dissertation presents an analysis of the representations of food biotechnologies in Italy. The thesis uses the analysis of discourse to illustrate the articulated ways in which representations are instantiated in different contexts. The theoretical thrust of the work resides in its discussion of the basic tenets of both Social Representations Theory and Discursive Psychology. The thesis offers a detailed description of the two frameworks; affinities and difference are highlighted, and a serious effort is made to develop an integrated set of theoretical resources to answer the research questions. The thesis proposes to combine a discursive methodology with Social Representations Theory. After a description of the relevant legislative framework follows an illustration of the categories used for the textual analysis. The study proposes the textual analysis of the following data: the first declaration issued by a small Italian council rejecting biotechnologies; four texts which focus on positions taken by the Catholic Church in the matter of food biotechnologies; several transcripts from a public debate in a small community of the north west of Italy. The latter study, which included an ethnographic dimension, focuses on recordings from interviews, a focus group, a public meeting and newspaper articles. Particular attention is paid to ideological representations and to the relevance of citizenship and governance to debates about food biotechnologies.

Explorations in the distributional and semantic similarity of words

A straightforward computation of the list of the words (the `tail words' of the list) that are distributionally most similar to a given word (the `head word' of the list) leads to the question: How semantically similar to the head word are the tail words; that is: how similar are their meanings to its meaning? And can we do better? The experiment was done on nearly 18,000 most frequent nouns in a Finnish newsgroup corpus. These nouns are considered to be distributionally similar to the extent that they occur in the same direct dependency relations with the same nouns, adjectives and verbs. The extent of the similarity of their computational representations is quantified with the information radius. The semantic classification of head-tail pairs is intuitive; some tail words seem to be semantically similar to the head word, some do not. Each such pair is also associated with a number of further distributional variables. Individually, their overlap for the semantic classes is large, but the trained classification-tree models have some success in using combinations to predict the semantic class. The training data consists of a random sample of 400 head-tail pairs with the tail word ranked among the 20 distributionally most similar to the head word, excluding names. The models are then tested on a random sample of another 100 such pairs. The best success rates range from 70% to 92% of the test pairs, where a success means that the model predicted my intuitive semantic class of the pair. This seems somewhat promising when distributional similarity is used to capture semantically similar words. This analysis also includes a general discussion of several different similarity formulas, arranged in three groups: those that apply to sets with graded membership, those that apply to the members of a vector space, and those that apply to probability mass functions.

Supistamista, rationalisointia vai kehittämistä? : Valtiovarainministeriön järjestelyosaston järjestämän henkilöstökoulutus- ja tutkimustoiminnan rakentuminen osaksi valtionhallintoa vuosina 1943-1971

The study examines the personnel training and research activities carried out by the Organization and Methods Division of the Ministry of Finance and their becoming a part and parcel of the state administration in 1943-1971. The study is a combination of institutional and ideological historical research in recent history on adult education, using a constructionist approach. Material salient to the study comes from the files of the Organization and Methods Division in the National Archives, parliamentary documents, committee reports, and the magazines. The concentrated training and research activities arranged by the Organization and Methods Division, became a part and parcel of the state administration in the midst of controversial challenges and opportunities. They served to solve social problems which beset the state administration as well as contextual challenges besetting rationalization measures, and organizational challenges. The activities were also affected by a dependence on decision-makers, administrative units, and civil servants organizations, by different views on rationalization and the holistic nature of reforms, as well as by the formal theories that served as resources. It chose long-term projects which extended to the political decision-makers and administrative units turf, and which were intended to reform the structures of the state administration and to rationalize the practices of the administrative units. The crucial questions emerged in opposite pairs (a constitutional state vs. the ideology of an administratively governed state, a system of national boards vs. a system of government through ministries, efficiency of work vs. pleasantness of work, centralized vs. decentralized rationalization activities) which were not solvable problems but impossible questions with no ultimate answers. The aim and intent of the rationalization of the state administration (the reform of the central, provincial, and local governments) was to facilitate integrated management and to render a greater amount of work by approaching management procedures scientifically and by clarifying administrative instances and their respon-sibilities in regards to each other. The means resorted to were organizational studies and committee work. In the rationalization of office work and finance control, the idea was to effect savings in administrative costs and to pare down those costs as well as to rationalize and heighten those functions by developing the institution of work study practitioners in order to coordinate employer and employee relationships and benefits (the training of work study practitioners, work study, and a two-tier work study practitioner organization). A major part of the training meant teaching and implementing leadership skills in practice, which, in turn, meant that the learning environment was the genuine work community and efforts to change it. In office rationalization, the solution to regulate the relations between the employer and the employees was the co-existence of the technical and biological rationalization and the human resource administration and the accounting and planning systems at the turn of the 1960s and 1970s. The former were based on the school of scientific management and human relations, the latter on system thinking, which was a combination of the former two. In the rationalization of the state administration, efforts were made to find solutions to stabilize management ideologies and to arrange the relationships of administrative systems in administrative science - among other things, in the Hoover Committee and the Simon decision making theory, and, in the 1960s, in system thinking. Despite the development-related vocabulary, the practical work was advanced rationalization. It was said that the practical activities of both the state administration and the administrative units depended on professional managers who saw to production results and human relations. The pedagogic experts hired to develop training came up with a training system, based on the training-technological model where the training was made a function of its own. The State Training Center was established and the training office of the Organization and Methods Division became the leader and coordinator of personnel training.

The Protection of Traditional Knowledge Associated with Genetic Resources: The Role of Databases and Registers

"The Protection of Traditional Knowledge Associated with Genetic Resources: The Role of Databases and Registers" ABSTRACT Yovana Reyes Tagle The misappropriation of TK has sparked a search for national and international laws to govern the use of indigenous peoples knowledge and protection against its commercial exploitation. There is a widespread perception that biopiracy or illegal access to genetic resources and associated traditional knowledge (TK) continues despite national and regional efforts to address this concern. The purpose of this research is to address the question of how documentation of TK through databases and registers could protect TK, in light of indigenous peoples increasing demands to control their knowledge and benefit from its use. Throughout the international debate over the protection of TK, various options have been brought up and discussed. At its core, the discussion over the legal protection of TK comes down to these issues: 1) The doctrinal question: What is protection of TK? 2) The methodological question: How can protection of TK be achieved? 3) The legal question: What should be protected? And 4) The policy questions: Who has rights and how should they be implemented? What kind of rights should indigenous peoples have over their TK? What are the central concerns the TK databases want to solve? The acceptance of TK databases and registers may bring with it both opportunities and dangers. How can the rights of indigenous peoples over their documented knowledge be assured? Documentation of TK was envisaged as a means to protect TK, but there are concerns about how documented TK can be protected from misappropriation. The methodology used in this research seeks to contribute to the understanding of the protection of TK. The steps taken in this research attempt to describe and to explain a) what has been done to protect TK through databases and registers, b) how this protection is taking place, and c) why the establishment of TK databases can or cannot be useful for the protection of TK. The selected case studies (Peru and Venezuela) seek to illustrate the complexity and multidisciplinary nature of the establishment of TK databases, which entail not only legal but also political, socio-economic and cultural issues. The study offers some conclusions and recommendations that have emerged after reviewing the national experiences, international instruments, work of international organizations, and indigenous peoples perspectives. This thesis concludes that if TK is to be protected from disclosure and unauthorized use, confidential databases are required. Finally, the TK database strategy needs to be strengthened by the legal protection of the TK itself.

Three-year-olds are sensitive to semantic prominence during online language comprehension

Recent evidence from adult pronoun comprehension suggests that semantic factors such as verb transitivity affect referent salience and thereby anap- hora resolution. We tested whether the same semantic factors influence pronoun comprehension in young children. In a visual world study, 3-year- olds heard stories that began with a sentence containing either a high or a low transitivity verb. Looking behaviour to pictures depicting the subject and object of this sentence was recorded as children listened to a subsequent sentence containing a pronoun. Children showed a stronger preference to look to the subject as opposed to the object antecedent in the low transitivity condition. In addition there were general preferences (1) to look to the subject in both conditions and (2) to look more at both potential antecedents in the high transitivity condition. This suggests that children, like adults, are affected by semantic factors, specifically semantic prominence, when interpreting anaphoric pronouns.

The RATAS-project, in the Light of 25 Years of Development Towards Computer-Integrated Construction

In smaller countries where the key players in construction IT development tend to know each other personally and where public R&D funding is concentrated to a few channels, IT roadmaps and strategies would seem to have a better chance of influencing development than in the bigger industrial countries. In this paper Finland and the RATAS-project is presented as a historical case illustrating such impact. RATAS was initiated as a construction IT roadmap project in 1985, involving many of the key organisations and companies active in construction sector development. Several of the individuals who took an active part in the project have played an important role in later developments both in Finland and on the international scene. The central result of RATAS was the identification of what is nowadays called Building Information Modelling (BIM) technology as the central issue in getting IT into efficient use in the construction sector. BIM, which earlier was referred to as building product modelling, has been a key ingredient in many roadmaps since and the subject of international standardisation efforts such as STEP and IAI/IFCs. The RATAS project can in hindsight be seen as a forerunner with an impact which also transcended national borders.

A Cointegration Approach to Topics in Empirical Macroeconomics (summary section only)

Mikael Juselius’ doctoral dissertation covers a range of significant issues in modern macroeconomics by empirically testing a number of important theoretical hypotheses. The first essay presents indirect evidence within the framework of the cointegrated VAR model on the elasticity of substitution between capital and labor by using Finnish manufacturing data. Instead of estimating the elasticity of substitution by using the first order conditions, he develops a new approach that utilizes a CES production function in a model with a 3-stage decision process: investment in the long run, wage bargaining in the medium run and price and employment decisions in the short run. He estimates the elasticity of substitution to be below one. The second essay tests the restrictions implied by the core equations of the New Keynesian Model (NKM) in a vector autoregressive model (VAR) by using both Euro area and U.S. data. Both the new Keynesian Phillips curve and the aggregate demand curve are estimated and tested. The restrictions implied by the core equations of the NKM are rejected on both U.S. and Euro area data. These results are important for further research. The third essay is methodologically similar to essay 2, but it concentrates on Finnish macro data by adopting a theoretical framework of an open economy. Juselius’ results suggests that the open economy NKM framework is too stylized to provide an adequate explanation for Finnish inflation. The final essay provides a macroeconometric model of Finnish inflation and associated explanatory variables and it estimates the relative importance of different inflation theories. His main finding is that Finnish inflation is primarily determined by excess demand in the product market and by changes in the long-term interest rate. This study is part of the research agenda carried out by the Research Unit of Economic Structure and Growth (RUESG). The aim of RUESG it to conduct theoretical and empirical research with respect to important issues in industrial economics, real option theory, game theory, organization theory, theory of financial systems as well as to study problems in labor markets, macroeconomics, natural resources, taxation and time series econometrics. RUESG was established at the beginning of 1995 and is one of the National Centers of Excellence in research selected by the Academy of Finland. It is financed jointly by the Academy of Finland, the University of Helsinki, the Yrjö Jahnsson Foundation, Bank of Finland and the Nokia Group. This support is gratefully acknowledged.