Scouts, Promoters, and Connectors: The Roles of Ratings in Nearest-Neighbor Collaborative Filtering

Recommender systems aggregate individual user ratings into predictions of products or services that might interest visitors. The quality of this aggregation process crucially affects the user experience and hence the effectiveness of recommenders in e-commerce. We present a characterization of nearest-neighbor collaborative filtering that allows us to disaggregate global recommender performance measures into contributions made by each individual rating. In particular, we formulate three roles-scouts, promoters, and connectors-that capture how users receive recommendations, how items get recommended, and how ratings of these two types are themselves connected, respectively. These roles find direct uses in improving recommendations for users, in better targeting of items and, most importantly, in helping monitor the health of the system as a whole. For instance, they can be used to track the evolution of neighborhoods, to identify rating subspaces that do not contribute ( or contribute negatively) to system performance, to enumerate users who are in danger of leaving, and to assess the susceptibility of the system to attacks such as shilling. We argue that the three rating roles presented here provide broad primitives to manage a recommender system and its community.

Uuden median murros Alma Mediassa, Sanoma Oy:ssä ja Yleisradiossa 1994-2004

This is a qualitative and multimethodological comparative study, which consists of two main parts: examining the development of new media and analysing and comparing the new media strategies of the three companies studied (Alma Media, Sanoma and the Finnish Broadcasting Company Yleisradio). The study includes the first large-scale review in Finnish of the development of new media, paying attention to the birth of the Internet as well as to mobile media, web TV and any other element of new media. It also concentrates on the function of electronic distribution channels before the age of the Internet, e.g. cable text and videotext. Answers about how the three traditional Finnish media houses began spreading their content to the Internet and wireless applications in 1994–2004 are also given. In researching the new media strategies the study pays special attention to the attitudes that the three media companies adopted towards the Internet and other forms of new media in their strategies during the years in question. By analysing and comparing, e.g., the companies’ strategies and their investments, the study ascertains whether the companies had a joint functional model in adopting new media or acted totally on their own without taking too much notice of the media field overall. The study makes extensive use of previously published material. The researcher has also interviewed almost twenty people who were involved in getting the companies’ new media functions under way. The methods for the interviews were dialogue and snowball sampling. The researcher has created a classification in which he divides the business strategies into four different categories: active strategy, careful strategy, permissive strategy, and passive strategy. In comparing and analysing the companies the researcher has used the classification devised by Allan Afuah & Christopher L. Tucci. The seven element classification consists of dominant managerial logic, competency trap, fear of cannibalisation and loss of revenue, channel conflict, political power, co-opetitor power and emotional attachment. In analysing the company strategies the researcher has also noted the classifications of convergence made by Everette E. Dennis and Graham Murdock as well as the aspects formulated by Sylvia Chan-Olmsted and Louisa Ha concerning the success of the companies in adopting the Internet into their functions. Based on all these classifications and by further developing them the researcher analyses and compares the success of the new media strategies of the three Finnish companies. The outcome of the study is a conclusion as to what kind of strategies the companies have carried out their new media functions and how they have succeeded in it.

A particle filtering approach for structural system identification in vehicle–structure interaction problems

The problem of identification of parameters of a beam-moving oscillator system based on measurement of time histories of beam strains and displacements is considered. The governing equations of motion here have time varying coefficients. The parameters to be identified are however time invariant and consist of mass, stiffness and damping characteristics of the beam and oscillator subsystems. A strategy based on dynamic state estimation method, that employs particle filtering algorithms, is proposed to tackle the identification problem. The method can take into account measurement noise, guideway unevenness, spatially incomplete measurements, finite element models for supporting structure and moving vehicle, and imperfections in the formulation of the mathematical models. Numerical illustrations based on synthetic data on beam-oscillator system are presented to demonstrate the satisfactory performance of the proposed procedure.

A new FIR filter for image restoration

Image filtering techniques have potential applications in biomedical image processing such as image restoration and image enhancement. The potential of traditional filters largely depends on the apriori knowledge about the type of noise corrupting the image. This makes the standard filters to be application specific. For example, the well-known median filter and its variants can remove the salt-and-pepper (or impulse) noise at low noise levels. Each of these methods has its own advantages and disadvantages. In this paper, we have introduced a new finite impulse response (FIR) filter for image restoration where, the filter undergoes a learning procedure. The filter coefficients are adaptively updated based on correlated Hebbian learning. This algorithm exploits the inter pixel correlation in the form of Hebbian learning and hence performs optimal smoothening of the noisy images. The application of the proposed filter on images corrupted with Gaussian noise, results in restorations which are better in quality compared to those restored by average and Wiener filters. The restored image is found to be visually appealing and artifact-free

Composite signal decomposition by digital inverse filtering

Inverse filters are conventionally used for resolving overlapping signals of identical waveshape. However, the inverse filtering approach is shown to be useful for resolving overlapping signals, identical or otherwise, of unknown waveshapes. Digital inverse filter design based on autocorrelation formulation of linear prediction is known to perform optimum spectral flattening of the input signal for which the filter is designed. This property of the inverse filter is used to accomplish composite signal decomposition. The theory has been presented assuming constituent signals to be responses of all-pole filters. However, the approach may be used for a general situation.

Kuluttajien kokemukset lastenvaatteiden ja -tarvikkeiden verkkokaupan palvelun laadusta sosiaalisen median keskusteluissa

Tämän pro gradu- tutkielman tavoitteena oli löytää kuluttajien palvelukokemukseen vaikuttavat tekijät lastenvaatteiden- ja tarvikkeiden -verkkokaupassa. Empiirinen tutkimus lähestyi aihetta netnografian avulla. Aineistoksi tähän tutkielmaan valittiin sosiaalisen median keskustelut. Tutkimuksen teoreettinen osa muodostui kolmesta aiheesta: verkko-ostamisesta ja palvelun laadusta, sosiaalisesta mediasta sekä word-of-mouth-viestinnästä. Verkko-ostamisessa käsiteltiin ostamisen eri vaiheet, kuluttajien motivaatiot verkosta ostamiseen, sekä siihen liittyvät riskit ja tarkasteltiin sähköisen palvelun laatua. Sosiaalinen media-luku kertoi sosiaalisen median käytöstä sekä eri medioista. Word-of-mouth-viestintä esitteli perinteisen word-of-mouth-viestinnän lisäksi sähköisen WOM-viestinnän ulottuvuudet. Empiirinen tutkimus oli laadullinen ja se toteutettiin netnografisesti. Netnografia on etnografiaan perustuva menetelmä, jota käytetään internet-aineistoissa. Aineiston analysoinnissa käytettiin teemoittelua. Aihetta lähestyttiin faktanäkökulmasta, eli tutkimuksessa oltiin kiinnostuneita niistä tosiasioista joita keskustelijat kertoivat. Aineisto kerättiin perhe-aiheisten lehtien keskustelupalstoilta sekä blogeista Google-haun avulla. Sosiaalisen median keskustelut valittiin aineistoksi, koska niistä uskottiin saatavan kaikkein totuudenmukaisinta tietoa, johon tutkijan kysymyksen asettelu ei vaikuta. Spontaanit keskustelut antavat erilaista tietoa kuin suoran kysymyksen vastaukset. Tutkimuksen tuloksena löytyi seitsemän teemaa, joita keskusteluissa käsiteltiin. Nämä ovat toimitus, palvelu, palautus ja normalisointi, ulkoasu ja toimivuus, hinta, maksaminen sekä tuotteet ja valikoima. Sen sijaan teorian pohjalta odotettavissa olleet turvallisuus ja yksityisyys eivät tulleet aineistosta lainkaan esiin. Erityisen huonona palveluna asiakkaat pitivät varastosaldojen paikkansapitämättömyyttä, hitaita toimituksia sekä epäystävällistä palvelua. Hyvää palvelua olivat nopeat toimitukset sekä yksilöllinen palvelu ja reklamaatioiden hyvä hoito.

Sosiaalisen median odotuksia ja käyttötarpeita osana asiantuntijaorganisaation viestintää

Tarkastelen pro gradu -tutkielmassani työyhteisön viestintätapojen muutosta ja sosiaalisen mediai roolia muutoksessa. Tutkimukseni tavoitteena on jäsentää ja ymmärtää sosiaalisen median roolia työyhteisön viestinnässä ja tarkastella miten sosiaalinen media taipuu tietoperustaisen asiantuntiorganisaation tarpeisiin. Tutkimuksen teoreettisessa osuudessa selvitän sitä, miten työn tekemisessä tapahtuneet muutokset ovat muuttaneet työyhteisön tapaa ja tarvetta viestiä. Työn tekemisen uusia muotoja ja verkostoitumista jäsennän tutkimuskirjallisuuden Alvesson (2000), Gruber & Palonen (2007), Kasvio & Tjäder (2007), Sennett (2007) kautta. Työn uusia muotoja voidaan kuvata käsitteillä asiantuntijakeskeisyys, tietointensiivisyys ja tietoperustainen työ. Kuva tutkimuksessani myös sosiaalisen median käsitettä sekä sen tunnusomaisuuksia. Tarkastelen tietoperustaisen asiantuntijaorganisaation odotuksia ja käyttötarpeita sosiaaliselle medialle Elisa Juholinin (2008) määrittelemän työyhteisöviestinnän uuden agendan (agendamalli) viitekehykses Tutkimukseni empiirisessä osuudessa tarkastelen aihetta yhden tietoperustaisen asiantuntijaorganisaation jäsenten kautta. Tavoitteeni on selvittää 1) millaisia odotuksia asiantuntijaorganisaatiossa työskentelevillä on sosiaalisen median käytöstä osana työyhteisön viestintää sekä 2) millaisia käyttötarpeita asiantuntijaorganisaatiossa työskentelevillä on sosiaalis* median käytöstä osana työyhteisön viestintää. Tutkimusotteeni on laadullinen, ja aineiston keruur menetelmänä käytän fokusryhmäkeskustelua sekä teemahaastattelua. Analysoin aineiston teemoittelemalla. Aineiston koodauksessa käytin Atlas.ti -ohjelmaa. Tutkimukseni perusteella voidaan todeta, että tärkeimpänä sosiaalisen median odotuksiin ja käyttötarpeisiin liittyi ajatus käyttäjälähtöisyydestä ja vuorovaikutuksesta. Käyttäjälähtöisyys ja vuorovaikutus näkyivät erityisesti odotuksena viestinnän rutiinien, tiedon ja osaamisen jakamisena entistä avoimempaan ; vuorovaikutteisempaan suuntaan. Tutkimus antaa viitteitä siitä, että sosiaalisen median sovelluksilla on mahdollista tukea vuorovaikutusta juuri niissä työyhteisöviestinnän tehtävissä, joissa vuorovaikutus jää helposti vähäiseksi. Sosiaalisen median sovelluksia tarvitaan myös tietoperustaisen asiantuntijaorganisaation osaamisen ja ns. hiljaisen tiedonjakamiseen. Tutkimuksen perusteella sosiaalinen media osana työyhteisön viestintää taipuu tiedontarpeen tyydyttäjänä, tarjoaa vaihtoehdon viestintäkanaville ja viestinnän toimintamalleille, mahdollistaa osallistumaan ja vaikuttamaan työyhteisön asioihin sekä tarjoaa tavan koota työyhteisössä olevaa osaamista ja hiljaista tietoa.

Application of nonlinear filtering to credit risk

Merton's model views equity as a call option on the asset of the firm. Thus the asset is partially observed through the equity. Then using nonlinear filtering an explicit expression for likelihood ratio for underlying parameters in terms of the nonlinear filter is obtained. As the evolution of the filter itself depends on the parameters in question, this does not permit direct maximum likelihood estimation, but does pave the way for the `Expectation-Maximization' method for estimating parameters. (C) 2010 Elsevier B.V. All rights reserved.

Component wave delineation of ECG by filtering in the Fourier domain

A complete solution to the fundamental problem of delineation of an ECG signal into its component waves by filtering the discrete Fourier transform of the signal is presented. The set of samples in a component wave is transformed into a complex sequence with a distinct frequency band. The filter characteristics are determined from the time signal itself. Multiplication of the transformed signal with a complex sinusoidal function allows the use of a bank of low-pass filters for the delineation of all component waves. Data from about 300 beats have been analysed and the results are highly satisfactory both qualitatively and quantitatively.

Correlation filtering: a terrain correction method for aeromagnetic maps with application

In correlation filtering we attempt to remove that component of the aeromagnetic field which is closely related to the topography. The magnetization vector is assumed to be spatially variable, but it can be successively estimated under the additional assumption that the magnetic component due to topography is uncorrelated with the magnetic signal of deeper origin. The correlation filtering was tested against a synthetic example. The filtered field compares very well with the known signal of deeper origin. We have also applied this method to real data from the south Indian shield. It is demonstrated that the performance of the correlation filtering is superior in situations where the direction of magnetization is variable, for example, where the remnant magnetization is dominant.

Three- and Seven-Point Optimally Weighted Recursive Median Filters for Gas Turbine Diagnostics

Measured health signals incorporate significant details about any malfunction in a gas turbine. The attenuation of noise and removal of outliers from these health signals while preserving important features is an important problem in gas turbine diagnostics. The measured health signals are a time series of sensor measurements such as the low rotor speed, high rotor speed, fuel flow, and exhaust gas temperature in a gas turbine. In this article, a comparative study is done by varying the window length of acausal and unsymmetrical weighted recursive median filters and numerical results for error minimization are obtained. It is found that optimal filters exist, which can be used for engines where data are available slowly (three-point filter) and rapidly (seven-point filter). These smoothing filters are proposed as preprocessors of measurement delta signals before subjecting them to fault detection and isolation algorithms.

Artifact-free reconstruction from off-axis digital holograms through nonlinear filtering

We present experimental investigation of a new reconstruction method for off-axis digital holographic microscopy (DHM). This method effectively suppresses the object auto-correlation, commonly called the zero-order term, from holographic measurements, thereby suppressing the artifacts generated by the intensities of the two beams employed for interference from complex wavefield reconstruction. The algorithm is based on non-linear filtering, and can be applied to standard DHM setups, with realistic recording conditions. We study the applicability of the technique under different experimental configurations, such as topographic images of microscopic specimens or speckle holograms.

Noise and outlier removal from jet engine health signals using weighted FIR median hybrid filters

The removal of noise and outliers from measurement signals is a major problem in jet engine health monitoring. Topical measurement signals found in most jet engines include low rotor speed, high rotor speed. fuel flow and exhaust gas temperature. Deviations in these measurements from a baseline 'good' engine are often called measurement deltas and the health signals used for fault detection, isolation, trending and data mining. Linear filters such as the FIR moving average filter and IIR exponential average filter are used in the industry to remove noise and outliers from the jet engine measurement deltas. However, the use of linear filters can lead to loss of critical features in the signal that can contain information about maintenance and repair events that could be used by fault isolation algorithms to determine engine condition or by data mining algorithms to learn valuable patterns in the data, Non-linear filters such as the median and weighted median hybrid filters offer the opportunity to remove noise and gross outliers from signals while preserving features. In this study. a comparison of traditional linear filters popular in the jet engine industry is made with the median filter and the subfilter weighted FIR median hybrid (SWFMH) filter. Results using simulated data with implanted faults shows that the SWFMH filter results in a noise reduction of over 60 per cent compared to only 20 per cent for FIR filters and 30 per cent for IIR filters. Preprocessing jet engine health signals using the SWFMH filter would greatly improve the accuracy of diagnostic systems. (C) 2002 Published by Elsevier Science Ltd.