RACK: RApid Clustering using K-means algorithm

The k-means algorithm is an extremely popular technique for clustering data. One of the major limitations of the k-means is that the time to cluster a given dataset D is linear in the number of clusters, k. In this paper, we employ height balanced trees to address this issue. Specifically, we make two major contributions, (a) we propose an algorithm, RACK (acronym for RApid Clustering using k-means), which takes time favorably comparable with the fastest known existing techniques, and (b) we prove an expected bound on the quality of clustering achieved using RACK. Our experimental results on large datasets strongly suggest that RACK is competitive with the k-means algorithm in terms of quality of clustering, while taking significantly less time.

Intent Based Clustering of Search Engine Query Log

The keyword based search technique suffers from the problem of synonymic and polysemic queries. Current approaches address only theproblem of synonymic queries in which different queries might have the same information requirement. But the problem of polysemic queries,i.e., same query having different intentions, still remains unaddressed. In this paper, we propose the notion of intent clusters, the members of which will have the same intention. We develop a clustering algorithm that uses the user session information in query logs in addition to query URL entries to identify cluster of queries having the same intention. The proposed approach has been studied through case examples from the actual log data from AOL, and the clustering algorithm is shown to be successful in discerning the user intentions.

Scanning Tunneling Spectroscopy on Pr0.68Pb0.32MnO3 single crystals

Scanning tunneling microscopy/spectroscopy studies were carried out on single crystals of colossal magnetoresistive manganite Pr0. 68Pb0.32MnO3 at different temperatures in order to probe their spatial homogeneity across the metal-insulator transition temperature TM-I(similar to 255 K). A metallic behavior of the local conductance was observed for temperatures T < TM-I. Zero bias conductance (dI/dV)v=(0), which is directly proportional to the local surface density of states at the Fermi level, shows a single distribution at temperatures T < 200 K suggesting a homogeneous electronic phase at low temperatures. In a narrow temperature window of 200 K < T < TM-I, however, an inhomogeneous distribution of (dI/dV)v=(0) has been observed. This result gives evidence for phase separation in the transition region in this compound.

Pelon ja toivon välissä : Pätkätyöläisen urakerronta

This study analyzes the forming of the occupational identity of the well-educated fixed-term employees. Fixed-term employment contracts amongst the well-educated labour force are exceptionally common in Finland as compared to other European countries. Two groups of modern fixed-term employees are distinguished. The first comprises well-educated women employed in the public sector whose fixed-term employment often consists of successive periods as temporary substitutes. The other group comprises well-educated, upper white-collar men aged over 40, whose fixed-term employment careers often consist of jobs of project nature or posts that are filled for a fixed period only. Method of the study For the empirical data I interviewed 35 persons (26 women and 9 men) in 33 interviews, one of which was conducted by e-mail and one was a group interview. All the interviews were electronically recorded and coded. All the interviewees have two things in common: fixed-term employment and formal high education. Thirteen (13) of them are researchers, four nurses, four midwives, four journalists, and ten project experts. I used the snowball method to get in touch the interviewees. The first interviewees were those who were recommended by the trade unions and by my personal acquaintances. These interviewees, in turn, recommended other potential interviewees. In addition, announcements on the internet pages of the trade unions were used to reach other interviewees. In analysing process I read the research material several times to find the turning points in the narrative the interviewees told. I also searched for the most meaningful stories told and the meaning the interviewees gave to these stories and to the whole narrative. In addition to that I paid attention to co-production of the narrative with the interviewees and analyzed the narrative as performance to be able to search for the preferred identities the interviewees perform. (Riesman 2001, 698-701). I do not pay much attention to the question of truth of a narrative in the sense of its correspondence with facts; rather I think a working life narrative has two tasks: On the one hand one has to tell the facts and on the other hand, he/she has to describe the meaning of these facts to herself/himself. To emphasize the double nature of the narrative about one’s working life I analyzed the empirical data both by categorizing it according to the cultural models of storytelling (heroic story, comedy, irony and tragedy) and by studying the themes most of the interviewees talked about. Ethics of the study I chose to use narrative within qualitative interviews on the grounds that in my opinion is more ethical and more empowering than the more traditional structured interview methods. During the research process I carefully followed the ethical rules of a qualitative research. The purpose of the interviews and the research was told to the interviewees by giving them a written description of the study. Oral permission to use the interview in this research was obtained from the interviewees. The names and places, which are mentioned in the study, are changed to conceal the actual identity of the interviewees. I shared the analysis with the interviewees by sending each of them the first analysis of their personal interview. This way I asked them to make sure that the identity was hidden well enough and hoped to give interviewees a chance to look at their narratives, to instigate new actions and sustain the present one (Smith 2001, 721). Also I hoped to enjoy a new possibility of joint authorship. Main results As a result of the study I introduce six models of telling a story. The four typical western cultural models that guide the telling are: heroic story, comedy, tragedy and satirical story (Hänninen 1999). In addition to these models I found two ways of telling a career filled with fixed-term employments that differ significantly from traditional career story telling. However, the story models in which the interviewees pour their experience locates the fixed term employers work career in an imagined life trajectory and reveals the meaning they give to it. I analyze the many sided heroic story that Liisa tells as an example of the strength of the fear of failing or losing the job the fixed term employee feels. By this structure it is also possible to show that success is felt to be entirely a matter of chance. Tragedy, the failure in one’s trial to get something, is a model I introduce with the help of Vilppu’s story. This narrative gets its meaning both from the sorrow of the failure in the past and the rise of something new the teller has found. Aino tells her story as a comedy. By introducing her narrative, I suggest that the purpose of the comedy, a stronger social consensus, gets deeper and darker shade by fixed-term employment: one who works as a fixed term employee has to take his/her place in his/her work community by him/herself without the support the community gives to those in permanent position. By studying the satiric model Rauno uses, I argue that using irony both turns the power structures to a carnival and builds free space to the teller of the story and to the listener. Irony also helps in building a consensus, mutual understanding, between the teller and the listener and it shows the distance the teller tells to exist between him and others. Irony, however, demands some kind of success in one’s occupational career but also at least a minor disappointment in the progress of it. Helmi tells her story merely as a detective story. By introducing Helmi’s narrative, I argue that this story model strengthens the trust in fairness of the society the teller and the listener share. The analysis also emphasizes the central position of identity work, which is caused by fixed-term employment. Most of the interviewees talked about getting along in working life. I introduced Sari’s narrative as an example of this. In both of these latter narratives one’s personal character and habits are lifted as permanent parts of the actual professional expertise, which in turn varies according to different situations. By introducing these models, I reveal that the fixed-term employees have different strategies to cope with their job situations and these strategies vary according to their personal motives and situations and the actual purpose of the interview. However, I argue that they feel the space between their hopes and fears narrow and unsecure. In the research report I also introduce pieces of the stories – themes – that the interviewees use to build these survival strategies. They use their personal curriculum vitae or portfolio, their position in work community and their work morals to build their professional identity. Professional identity is flexible and varies in time and place, but even then it offers a tool to fix one’s identity work into something. It offers a viewpoint to society and a tool to measure one’s position in surrounding social nets. As one result of the study I analyze the position the fixed-term employees share on the edge of their job communities. I summarize the hopes and fears the interviewees have concerning employers, trade unions, educational institutions and the whole society. In their opinion, the solidarity between people has been weakened by the short-sighted power of the economy. The impact the fixed-term employment has on one’s professional identity and social capital is a many-sided and versatile process. Fixed-term employment both strengthens and weakens the professional identity, social capital and the building of trust. Fixed-term employment also affects one’s day-to-day life by excluding her/him from the norm and by one’s difficulty in making long-term plans (Jokinen 2005). Regardless of the nature of the job contract, the workers themselves are experts in making the best of their sometimes less than satisfying work life and they also build their professional identity by using creatively their education, work experiences and interpersonal relations. However, a long career of short fixed-term employments may seriously change the perception of employee about his/her role. He/she may start concentrating only in coping in his/her unsatisfactory situation and leaves the active improvement of the lousy working conditions to other people. Keywords: narrative, fixed-tem employment, occupational identity, work, story model, social capital, career  

Sensor network deployment for agronomical data gathering in semi-arid regions

We share our experience in planning, designing and deploying a wireless sensor network of one square kilometre area. Environmental data such as soil moisture, temperature, barometric pressure, and relative humidity are collected in this area situated in the semi-arid region of Karnataka, India. It is a hope that information derived from this data will benefit the marginal farmer towards improving his farming practices. Soon after establishing the need for such a project, we begin by showing the big picture of such a data gathering network, the software architecture we have used, the range measurements needed for determining the sensor density, and the packaging issues that seem to play a crucial role in field deployments. Our field deployment experiences include designing with intermittent grid power, enhancing software tools to aid quicker and effective deployment, and flash memory corruption. The first results on data gathering look encouraging.

Hyvinvointivaltion optimaalisen köyhyyspolitiikan määrittelyä

The purpose of this research is to identify the optimal poverty policy for a welfare state. Poverty is defined by income. Policies for reducing poverty are considered primary, and those for reducing inequality secondary. Poverty is seen as a function of the income transfer system within a welfare state. This research presents a method for optimising this function for the purposes of reducing poverty. It is also implemented in the representative population sample within the Income Distribution Data. SOMA simulation model is used. The iterative simulation process is continued until a level of poverty is reached at which improvements can no longer be made. Expenditures and taxes are kept in balance during the process. The result consists of two programmes. The first programme (social assistance programme) was formulated using five social assistance parameters, all of which dealt with the norms of social assistance for adults (€/month). In the second programme (basic benefits programme), in which social assistance was frozen at the legislative level of 2003, the parameter with the strongest poverty reduction effect turned out to be one of the basic unemployment allowances. This was followed by the norm of the national pension for a single person, two parameters related to housing allowance, and the norm for financial aid for students of higher education institutions. The most effective financing parameter measured by gini-coefficient in all programmes was the percent of capital taxation. Furthermore, these programmes can also be examined in relation to their costs. The social assistance programme is significantly cheaper than the basic benefits programme, and therefore with regard to poverty, the social assistance programme is more cost effective than the basic benefits programme. Therefore, public demand for raising the level of basic benefits does not seem to correspond to the most cost effective poverty policy. Raising basic benefits has most effect on reducing poverty within the group of people whose basic benefits are raised. Raising social assistance, on the other hand, seems to have a strong influence on the poverty of all population groups. The most significant outcome of this research is the development of a method through which a welfare state’s income transfer-based safety net, which has severely deteriorated in recent decades, might be mended. The only way of doing so involves either social assistance or some forms of basic benefits and supplementing these by modifying social assistance.

Kaupunkilaisen elinympäristöongelma : Käsitteellistämistavat ja puuttumisen mallit

The study analyses the reaction of urban residents to problems, i.e. disturbing factors, in their living environment, and also their ways of doing something about these problems. It is based on urban-sociological theory on everyday life in a modern metropolis. On this theoretical basis, problems in the urban living environment are analysed in terms of a policy of everyday interference: when urban citizens become aware of a problem in their environment, they face a pattern of behaviour where the norm is polite indifference and negative solidarity. They may feel they ought to do something about the problem, but at the same time, an implicit rule of urban life is not to interfere with other people s lives so they won t interfere with yours. For example, it is not that easy for someone disturbed by littering to complain directly to those who litter the streets. Or if you complain about tobacco smoke from the neighbour s balcony, your neighbours might get cross. Direct interference with a problem in the environment usually implies an encounter with a hitherto unknown counterpart and their possible counter-reaction. The risk is either to lose face or get into downright conflict. Therefore, an easier way may be to complain to the city authorities. The Helsinki City Environment Centre is currently working on solutions for all the various kinds of problems that occur in a dense urban structure. Various ways of conceptualising the problems in the living environment are analysed empirically using theme interviews made with citizens having contacted Helsinki City Environment Centre. A phenomenographic approach and a theory-based categorisation are applied on the analysis of the theme interviews. On the grounds of the analysis, the ways of conceptualising are determined by 1) the difficulty of interfering and convincing other people, which in practice means meddling in other people s business, 2) a territorial struggle for space and a place in a dense urban structure, 3) breaches of rules and norms for social routines in urban life, and 4) a crumbling of the urban identity and all that goes along with that. The analysis of the ways of conceptualisation is deepened using a cultural risk theory. The final outcome of the analysis is four types of behaviour among urban residents with regard to interference with everyday problems in the living environment. They have been called yard police , fence builder , park warden and environmental caretaker . The study combines an urban-sociological approach with the theoretical tradition of urban research and with research on municipal environmental policy.

Stadilaispojat, rikokset ja lastensuojelu : Viisi tapaustutkimusta kuudelta vuosikymmeneltä

Societal reactions to norm breaking behavior of children reveal, how we understand childhood, the relations between generations and communitie's ratio of tolerance. In Finland the children that repeatedly commit crimes receive social service measures that are based on Child Welfare Act. In the city of Helsinki (Stadi in the slang of Helsinki) existed an agency specifically established for ill-behaving children until the 1980's, agter which an unified agency for the maltreated and maladjusted children was founded. Through five boys' welfare cases, this research aims at defining what kind of positions, social relations and structures are constructed in the social dynamics of these children's everyday lives. The cases cover different decades from the 1940s to the present. At the same time the cases reflect the child welfare and societal practices, and reveal how the communities have participated in constructing deviance in different eras. The research is meta-theoretically based on critical realism and specifically on Roy Bhaskar's transformative model of social activity. The cases are analyzed in the framework of Edwin M. Lemert's societal reaction theory. Thus the focus of the study is on the wide structural context of the institutional and societal definitions of deviance. The research is methodologically based on a qualitative multiple case study research. The primary data consist of classified child welfare case files collected from the archives of the city of Helsinki. The data of the institutional level consist of the annual reports from 1943 to 2004 and the ordinances from 1907 onwards, and of various committee documents produced in the law-making process of child welfare, youth and criminal legislation of the 20th century. Empirical finding are interpreted in a dialogue with previous historical and child welfare research, contemporary literature and studies on the urban development. The analysis is based on Derek Layder's model of adaptive theory. The research forms a viewpoint to the historical study of child welfare, in which the historical era, its agents and the dynamics of their mutual relations are studied through an individual level reconstruction based on the societal reaction theory. The case analyses reveal how the positions of the children form differently in the different eras of child welfare practices. In the 1940s the child is positioned as a psychopath and a criminal type. The measures are aimed at protecting the community from the disturbed child, and at adjusting the individual by isolation. From 1960s to 1980s the child is positioned as a child in need of help and support. The child becomes a victim, a subject that occupies rights, and a target of protection. In the turn of the millennium a norm breaking child is positioned as a dangerous individual that, in the name of the community safety, has to be confined. The case analyses also reveal the prevailing academic and practical paradigms of the time. Keywords: childhood, youth, child protection, child welfare, delinquency, crime, deviance, history, critical realism, case study research

Basic Human Values in the Workplace

Työntekijöiden henkilökohtaisia arvoja ja niiden yhteyksiä asenteisiin ei ole juuri tutkittu. Tämän tutkimuksen tavoitteena oli selvittää, onko suomalaisessa metalliteollisuuden yrityksen henkilöstön (N=1314) arvojen rakenne S. H. Schwartzin arvoteorian mukainen. Lisäksi tutkittiin arvojen yhteyksiä organisaatiomuutosta koskeviin asenteisiin ja tiedon jakamiseen työyhteisössä. Arvomittarina käytettiin uutta 40-osioista Portrait Value Questionnairea (PVQ). Mittarin validiteetti osoitettiin ver-taamalla nyt kerätyn aineiston arvorakennetta aikaisemmalla mittarilla kerättyihin arvoteorian mukaisiin yliopisto-opiskelijoiden vastauksiin. Organisaatiomuutosta koskevien asenteiden ja tiedonjakamisen mittarit luotiin laadullisissa esitutkimuksissa. Tilastolliset analyysit osoittivat, että toimihenkilöiden ja työntekijöiden arvojen rakenteet noudattivat pääosin Schwartzin teoriaa, mutta turvallisuusarvot sijaitsivat molemmissa ryhmissä universalismin ja hyväntahtoisuuden joukossa. Universalismi ja hyväntahtoisuus ennustivat myönteistä asennetta organisaatiomuutoksia kohtaan, mutta perinteiden ja mielihyvän arvostaminen liittyivät kielteisiin muutosasenteisiin. Sosiaalisia normeja kunnioittavien eli yhdenmukaisuutta arvostavien henkilöiden muut arvot vaikuttivat muutosasenteisiin vähemmän kuin niillä, joille yhdenmukaisuus ei ollut tärkeää. Lisäksi suoriutumisarvon yhteys muutosasenteisiin oli yhdenmukaisuutta arvostavilla henkilöillä positiivinen, mutta niillä, jotka eivät arvostaneet yhdenmukaisuutta, yhteys oli negatiivinen. Itseohjautuvuutta arvostavat henkilöt pitivät työyhteisönsä tiedon jakamista heikompana, kun taas hyväntahtoisuutta ja yhdenmukaisuutta arvostavat pitivät sitä muihin nähden parempana. Suoriutumisarvo oli yhteydessä tiedonjakamiseen vain silloin, kun yhdenmukaisuus oli tärkeää. Työpaikkojen (N=19) keskiarvoja vertailtaessa havaittiin, että ne työpaikat, joissa arvostettiin paljon universalismia, hyväntahtoisuutta ja yhdenmukaisuutta sekä vähän valtaa ja suoriutumista saivat henkilöstöltään parhaat arvioinnit tiedon jakamisesta. Tutkimukseen osallistuneet henkilöt jaettiin työtehtäviensä perusteella kolmeen ammatilliseen ympäristöön: konven-tionaaliseen (mm. taloushallinto), realistiseen (mm. tuotanto) ja yrittäjämäiseen (mm. myynti). Yrittäjämäisessä ammatillisessa ympäristössä toimivat arvostivat enemmän kuin konventionaalisessa ympäristössä toimivat valtaa, itseohjautuvuutta ja suoriutumista. Realistisessa ympäristössä arvostettiin enemmän perinteitä ja mielihyvää kuin yrittäjämäisessä ympäristössä. Ryhmien väliset erot arvoissa johtuivat koulutuksesta, iästä ja sukupuolijakaumasta.

Exploring polymorphism by solvent mediation in potentially active herbicide Metribuzin: A subtle interplay of weak intermolecular interactions

Metribuzin, 4-amino-6-tert-butyl-3-methylthio- 1,2,4-triazin-5-one, exhibits polymorphic behaviour, crystallizing as plates and needles, driven by variation in solvent polarity, a delicate balance of weak intermolecular forces generating different molecular assemblies.

Hybrid learning scheme for data mining applications

Classification of large datasets is a challenging task in Data Mining. In the current work, we propose a novel method that compresses the data and classifies the test data directly in its compressed form. The work forms a hybrid learning approach integrating the activities of data abstraction, frequent item generation, compression, classification and use of rough sets.

Hybrid learning scheme for data mining applications

Classification of large datasets is a challenging task in Data Mining. In the current work, we propose a novel method that compresses the data and classifies the test data directly in its compressed form. The work forms a hybrid learning approach integrating the activities of data abstraction, frequent item generation, compression, classification and use of rough sets.

A silicon micromachined photonic band gap cell, characteristics and an application

A novel PBG cell based on micromachining of Silicon using wet anisotropic etching has been considered. Since this is based on etching of the Silicon substrate, it is amenable to fabrication with standard Silicon processes and integration with millimeter wave circuits. We characterize this kind of PBG cell by full wave simulations using a time domain code. For the purpose of characterization, the scenario of a 50 ohm microstrip line placed on a Silicon substrate which is anisotropically etched to create patterns with sloping walls is considered. This is shown to produce the well known PBG response of stop bands in certain frequency bands. We look at the variation in the transmission coefficient (S-21) response as the number of periods, length based average fill factor and depth of micromachining are varied. One application of a low pass filter has been proposed and simulated results are given.

Sequence design for symbol-asynchronous CDMA with power or rate constraints

Sequence design and resource allocation for a symbol-asynchronous chip-synchronous code division multiple access (CDMA) system is considered in this paper. A simple lower bound on the minimum sum-power required for a non-oversized system, based on the best achievable for a non-spread system, and an analogous upper bound on the sum rate are first summarised. Subsequently, an algorithm of Sundaresan and Padakandla is shown to achieve the lower bound on minimum sum power (upper bound on sum rate, respectively). Analogous to the synchronous case, by splitting oversized users in a system with processing gain N, a system with no oversized users is easily obtained, and the lower bound on sum power (upper bound on sum rate, respectively) is shown to be achieved by using N orthogonal sequences. The total number of splits is at most N - 1.

Sequential split vector quantization of LSF parameters using conditional PDF

A better performing product code vector quantization (VQ) method is proposed for coding the line spectrum frequency (LSF) parameters; the method is referred to as sequential split vector quantization (SeSVQ). The split sub-vectors of the full LSF vector are quantized in sequence and thus uses conditional distribution derived from the previous quantized sub-vectors. Unlike the traditional split vector quantization (SVQ) method, SeSVQ exploits the inter sub-vector correlation and thus provides improved rate-distortion performance, but at the expense of higher memory. We investigate the quantization performance of SeSVQ over traditional SVQ and transform domain split VQ (TrSVQ) methods. Compared to SVQ, SeSVQ saves 1 bit and nearly 3 bits, for telephone-band and wide-band speech coding applications respectively.