12. Artikkelit / Articles

Permanent URI for this communityhttps://aaltodoc.aalto.fi/handle/123456789/79

Arkisto - Ei ole enää käytössä. Tässä kokoelmassa on rinnakkaistallennettuja artikkeleita vuoteen 2022 asti. Uudet rinnakkaistallennetut artikkelit löytyvät Aalto-yliopiston tutkimustietojärjestelmästä.
Archive - No longer in use. This collection contains green open access articles up until the year 2022. New green open access articles can be found in Aalto University’s research information system.

Browse

Browsing 12. Artikkelit / Articles by Department "Department of Computer Science"

Filter results by typing the first few letters
Now showing 1 - 15 of 15
  • The current duration design for estimating the time to pregnancy distribution: a nonparametric Bayesian perspective
    (2015) Gasbarra, Dario; Arjas, Elja; Vehtari, Aki; Slama, Rémy; Keiding, Niels
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    This paper was inspired by the studies of Niels Keiding and co-authors on estimating the waiting time-to-pregnancy (TTP) distribution, and in particular on using the current duration design in that context. In this design, a cross-sectional sample of women is collected from those who are currently attempting to become pregnant, and then by recording from each the time she has been attempting. Our aim here is to study the identifiability and the estimation of the waiting time distribution on the basis of current duration data. The main difficulty in this stems from the fact that very short waiting times are only rarely selected into the sample of current durations, and this renders their estimation unstable. We introduce here a Bayesian method for this estimation problem, prove its asymptotic consistency, and compare the method to some variants of the non-parametric maximum likelihood estimators, which have been used previously in this context. The properties of the Bayesian estimation method are studied also empirically, using both simulated data and TTP data on current durations collected by Slama et al. (Hum Reprod 27(5):1489–1498, 2012).
  • Depth Artifacts Caused by Spatial Interlacing in Stereoscopic 3D Displays
    (2015) Hakala, Jussi H.; Oittinen, Pirkko; Häkkinen, Jukka P.
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
  • Developing ethical and transparent artificial intelligence algorithms to support decision making in healthcare based on brain research and personal care events of patients
    (2022) Lahti, Lauri
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    We propose a new research methodology that develops ethical and transparent artificial intelligence algorithms to support decision making in healthcare. This development relies on a diverse statistical and data analysis methodology based on real-life data gathered in brain research and care events of different patient groups. The proposed new research methodology is created, developed and carried out in a broad international multidisciplinary research collaboration with various patient and disabled people's groups, healthcare professionals, educational institutions, and laboratory measurements of experimental brain research conducted at a biomedical research institute. The proposed new research methodology is motivated by the previous research that has given successful classification results with various bio-inspired artificial intelligence algorithms, based on unsupervised learning (such as various clustering algorithms) and supervised learning (such as artificial neural network algorithms) that are implemented following the structural and functional principles of real-life living biological tissues.
  • Development of reasoning rules for artificial intelligence openly for all: method and results 20240422
    (2024) Lahti, Lauri
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    In this current research article manuscript we report now a new method and results that we have conceived, developed and created to enable a web service and its related research and development project that develop reasoning rules for a new artificial intelligence service openly in collaboration with different persons ("Development of reasoning rules for artificial intelligence openly for all"). Thus it is aimed to address as diversely as possible different persons and their different life situations, needs and wishes. Each phase of the development of reasoning rules is archived, and gathered data and results are published openly to be evaluated by everyone, at the same time taking care situationally about appropriate anonymization/pseudonymization of information. The diversity of participating persons and the publishing of details advance the ethicality of the artificial intelligence service as well as openness, trustworthiness and transparency. With the web service the person participates in its using, development and research and its related research and development project. The reasoning rules are developed with the web service so that the person is asked to answer to different interpretation tasks and inquiries of background information, and then the gathered data is analyzed to identify patterns and dependencies for example in respect to the health condition, care planning and life situation. Privacy notice and terms of use are presented on the starting page of the web service. To our best knowledge our research is the first of its kind to conceive, develop, create, use and publish the new method and results of gathering answers to interpretation tasks to enable development of reasoning rules for artificial intelligence openly for all in this current research article manuscript under the license Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International (CC BY-NC-ND 4.0).
  • Dynamic retrospective filtering of physiological noise in BOLD fMRI: DRIFTER
    (2012) Särkkä, Simo; Solin, Arno; Nummenmaa, Aapo; Vehtari, Aki; Auranen, Toni; Vanni, Simo; Lin, Fa-Hsuan
     School of Electrical Engineering | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    In this article we introduce the DRIFTER algorithm, which is a new model based Bayesian method for retrospective elimination of physiological noise from functional magnetic resonance imaging (fMRI) data. In the method, we first estimate the frequency trajectories of the physiological signals with the interacting multiple models (IMM) filter algorithm. The frequency trajectories can be estimated from external reference signals, or if the temporal resolution is high enough, from the fMRI data. The estimated frequency trajectories are then used in a state space model in combination of a Kalman filter (KF) and Rauch–Tung–Striebel (RTS) smoother, which separates the signal into an activation related cleaned signal, physiological noise, and white measurement noise components. Using experimental data, we show that the method outperforms the RETROICOR algorithm if the shape and amplitude of the physiological signals change over time.
  • Enabling personalized healthcare by analyzing semantic dependencies in a conceptual co-occurrence network based on a medical vocabulary
    (2016) Lahti, Lauri
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    The amount of medical knowledge is constantly growing thus providing new hope for people having health-related problems. However a challenge is to develop flexible methods to facilitate managing and interpreting large medical knowledge entities. There is a need to enhance health literacy by developing personalized health support tools. Furthermore there is a need to assist decision-making with decision support tools. The recent and on-going changes in everyday life both on technological and societal levels (for example adoption of smart phones and personal mobile medical tracking devices, social networking, open source and open data initiatives, fast growth of accumulated medical data, need for new self-care solutions for aging European population) motivate to invest in the development of new computerized personalized methods for knowledge management of medical data for diagnosis and treatment. To enable creation of new adaptive personalized health support tools we have carried out an evaluation of semantic dependencies in a conceptual co-occurrence network covering a set of concepts of a medical vocabulary with experimental results ranging up to 2994 unique nouns, 82814 unique conceptual links and 200000 traversed link steps.
  • Hierarchical second-order analysis of replicated spatial point patterns with non-spatial covariates
    (2014) Myllymäki, Mari; Särkkä, Aila; Vehtari, Aki
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    In this paper we propose a method for incorporating the effect of non-spatial covariates into the spatial second-order analysis of replicated point patterns. The variance stabilizing transformation of Ripley’s K function is used to summarize the spatial arrangement of points, and the relationship between this summary function and covariates is modelled by hierarchical Gaussian process regression. In particular, we investigate how disease status and some other covariates affect the level and scale of clustering of epidermal nerve fibres. The data are point patterns with replicates extracted from skin blister samples taken from 47 subjects.
  • How players across gender and age experience Pokémon Go?
    (2019) Malik, Aqdas; Hiekkanen, Kari; Hussain, Zaheer; Hamari, Juho; Johri, Aditya
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    The purpose of this study is to provide insights into player experiences and motivations in Pokémon Go, a relatively new phenomenon of location-based augmented reality games. With the increasing usage and adoption of various forms of digital games worldwide, investigating the motivations for playing games has become crucial not only for researchers but for game developers, designers, and policy makers. Using an online survey (N = 1190), the study explores the motivational, usage, and privacy concerns variations among age and gender groups of Pokémon Go players. Most of the players, who are likely to be casual gamers, are persuaded toward the game due to nostalgic association and word of mouth. Females play Pokémon Go to fulfill physical exploration and enjoyment gratifications. On the other hand, males seek to accomplish social interactivity, achievement, coolness, and nostalgia gratifications. Compared to females, males are more concerned about the privacy aspects associated with the game. With regard to age, younger players display strong connotation with most of the studied gratifications and the intensity drops significantly with an increase in age. With the increasing use of online and mobile games worldwide among all cohorts of society, the study sets the way for a deeper analysis of motivation factors with respect to age and gender. Understanding motivations for play can provide researchers with the analytic tools to gain insight into the preferences for and effects of game play for different kinds of users.
  • Intelligent Products: Shifting the Production Control Logic in Construction (With Lean and BIM)
    (2015) Dave, Bhargav; Kubler, Silvain; Pikas, Ergo; Holmström, Jan; Singh, Vishal; Främling, Kary; Koskela, Lauri
     School of Science | A4 Artikkeli konferenssijulkaisussa
    Production management and control in construction has not been addressed/updated ever since the introduction of Critical Path Method and the Last Planner® system. The predominant outside-in control logic and a fragmented and deep supply chain in construction significantly affect the efficiency over a lifecycle. In a construction project, a large number of organisations interact with the product throughout the process, requiring a significant amount of information handling and synchronisation between these organisations. However, due to the deep supply chains and problems with lack of information integration, the information flow down across the lifecycle poses a significant challenge. This research proposes a product centric system, where the control logic of the production process is embedded within the individual components from the design phase. The solution is enabled by a number of technologies and tools such as Building Information Modelling, Internet of Things, Messaging Systems and within the conceptual process framework of Lean Construction. The vision encompasses the lifecycle of projects from design to construction and maintenance, where the products can interact with the environment and its actors through various stages supporting a variety of actions. The vision and the tools and technologies required to support it are described in this paper.
  • Interpretation of health-related expressions and dialogues: enabling personalized care with contextual measuring and machine learning
    (2017) Lahti, Lauri
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    We propose a new research framework that develops a method for interpretation of health-related expressions and dialogues to enable personalized care with contextual measuring and machine learning. The new research framework is implemented with a research project that gathers from various patient groups and other population groups a broad collection of essential perspectives towards health and well-being. In experimental setups persons (for example patients, their family members and representatives of care personnel) are asked to classify a given set of expressions (linguistic statements, image materials or other stimuli) into different categories, and these categorizations are then used as input vectors for computational models. To develop the method a central task is to classify with machine learning models health-related expressions and dialogues in respect to various events, processes and persons in healthcare. Our experimental results based on a sample of context-based linguistic health data indicated fruitful possibilities for gaining classifications of essential traits of language usage, appearance and activity for persons of diverse population groups based on various scales, perspectives, background assumptions and contexts.
  • Know Your Phish: Novel Techniques for Detecting Phishing Sites and Their Targets
    (2016) Marchal, Samuel; Saari, Kalle; Singh, Nidhi; Asokan, N.
     School of Science | A4 Artikkeli konferenssijulkaisussa
    Phishing is a major problem on the Web. Despite the significant attention it has received over the years, there has been no definitive solution. While the state-of-the-art solutions have reasonably good performance, they require a large amount of training data and are not adept at detecting phishing attacks against new targets. In this paper, we begin with two core observations: (a) although phishers try to make a phishing webpage look similar to its target, they do not have unlimited freedom in structuring the phishing webpage, and (b) a webpage can be characterized by a small set of key terms, how these key terms are used in different parts of a webpage is different in the case of legitimate and phishing webpages. Based on these observations, we develop a phishing detection system with several notable properties: it requires very little training data, scales well to much larger test data, is language-independent, fast, resilient to adaptive attacks and implemented entirely on client-side. In addition, we developed a target identification component that can identify the target website that a phishing webpage is attempting to mimic. The target detection component is faster than previously reported systems and can help minimize false positives in our phishing detection system.
  • Off-the-Hook: An Efficient and Usable Client-Side Phishing Prevention Application
    (2017) Marchal, Samuel; Armano, Giovanni; Grondahl, Tommi; Saari, Kalle; Singh, Nidhi; Asokan, N.
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    Phishing is a major problem on the Web. Despite the significant attention it has received over the years, there has been no definitive solution. While the state-of-the-art solutions have reasonably good performance, they suffer from several drawbacks including potential to compromise user privacy, difficulty of detecting phishing websites whose content change dynamically, and reliance on features that are too dependent on the training data. To address these limitations we present a new approach for detecting phishing webpages in real-time as they are visited by a browser. It relies on modeling inherent phisher limitations stemming from the constraints they face while building a webpage. Consequently, the implementation of our approach, Off-the-Hook, exhibits several notable properties including high accuracy, brand-independence and good language-independence, speed of decision, resilience to dynamic phish and resilience to evolution in phishing techniques. Off-the-Hook is implemented as a fully-client-side browser add-on, which preserves user privacy. In addition, Off-the-Hook identifies the target website that a phishing webpage is attempting to mimic and includes this target in its warning. We evaluated Off-the-Hook in two different user studies. Our results show that users prefer Off-the-Hook warnings to Firefox warnings.
  • Sharper Upper Bounds for Unbalanced Uniquely Decodable Code Pairs
    (2018) Austrin, Per; Kaski, Petteri; Koivisto, Mikko; Nederlof, Jesper
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    Two sets of 0-1 vectors of fixed length form a uniquely decodeable code pair if their Cartesian product is of the same size as their sumset, where the addition is pointwise over integers. For the size of the sumset of such a pair, van Tilborg has given an upper bound in the general case. Urbanke and Li, and later Ordentlich and Shayevitz, have given better bounds in the unbalanced case, that is, when either of the two sets is sufficiently large. Improvements to the latter bounds are presented.
  • Supporting care by interpretation of expressions about patient experience with machine learning
    (2018) Lahti, Lauri
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    Our research aims at addressing the needs of developing data analysis about communication in respect to care seeking and primary care, discovering how health expressions evolve along the personal growth and learning process, and how to solve the needs identified in respect to developing measuring the quality of life. We provide an overview of the development of a new research methodology exploiting machine learning for analyzing patient experience expressions to support personalized care and managing in everyday life. Our research relies on an online questionnaire in which the representatives of various population groups perform interpretation tasks. Dependencies between answers about the interpretation tasks and background information are analyzed with machine learning methods. The research creates new ways to interpret and address the meanings of language usage of different groups of patients and impaired carefully and distinctively as a part of everyday life and care events.
  • Uses and Gratifications of digital photo sharing on Facebook
    (2015) Malik, Aqdas; Dhir, Amandeep; Nieminen, Marko
     School of Science | A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
    Despite the rapid adoption of Facebook as a means of photo sharing, minimal research has been conducted to understand user gratification behind this activity. In order to address this gap, the current study examines users’ gratifications in sharing photos on Facebook by applying Uses and Gratification (U&G) theory. An online survey completed by 368 respondents identified six different gratifications, namely, affection, attention seeking, disclosure, habit, information sharing, and social influence, behind sharing digital photos on Facebook. Some of the study’s prominent findings were: age was in positive correlation with disclosure and social influence gratifications; gender differences were identified among habit and disclosure gratifications; number of photos shared was negatively correlated with habit and information sharing gratifications. The study’s implications can be utilized to refine existing and develop new features and services bridging digital photos and social networking services.