A5064247235- Semantic Web and Ontologies
- Information Retrieval and Search Behavior
- EEG and Brain-Computer Interfaces
- Personal Information Management and User Behavior
- Mobile Crowdsensing and Crowdsourcing
- Recommender Systems and Techniques
- Advanced Text Analysis Techniques
- Topic Modeling
- Expert finding and Q&A systems
- Data Visualization and Analytics
- Image Retrieval and Classification Techniques
- Neural and Behavioral Psychology Studies
- Service-Oriented Architecture and Web Services
- Advanced Database Systems and Queries
- Neural dynamics and brain function
- Data Quality and Management
- Biomedical Text Mining and Ontologies
- Web Data Mining and Analysis
- Face Recognition and Perception
- Data Management and Algorithms
- Advanced Chemical Sensor Technologies
- Advanced Image and Video Retrieval Techniques
- Emotion and Mood Recognition
- Visual Attention and Saliency Detection
- Functional Brain Connectivity Studies
Lappeenranta-Lahti University of Technology
2023-2025
University of Copenhagen
2021-2025
University of Helsinki
2009-2025
IT University of Copenhagen
2024
Finland University
2023
Helsinki Institute for Information Technology
2012-2020
Aalto University
2007-2019
University of California, Berkeley
2012-2013
Wearable devices that measure and record physiological signals are now becoming widely available to the general public with ever-increasing affordability signal quality. The data from these introduce serious ethical challenges remain largely unaddressed. Users do not always understand how can be leveraged reveal private information about them developers of may fully grasp collected today could used in future for completely different purposes. We discuss potential wearable devices, initially...
The system should let users incrementally direct their search toward relevant, though not initially obvious, information.
Techniques for both exploratory and known item search tend to direct only more specific subtopics or individual documents, as opposed allowing directing the exploration of information space. We present an interactive retrieval system that combines Reinforcement Learning techniques along with a novel user interface design allow active engagement users in search. Users can directly manipulate document features (keywords) indicate their interests is used model by trade off between exploitation....
Fairness is an emerging and challenging topic in recommender systems. In recent years, various ways of evaluating therefore improving fairness have emerged. this study, we examine existing evaluation measures Specifically, focus solely on exposure-based individual items that aim to quantify the disparity how are recommended users, separate from item relevance users. We gather all such critically analyse their theoretical properties. identify a series limitations each them, which collectively...
This article presents the vision and results of creating basis for a national semantic Web content infrastructure in Finland 2003-2007. The main elements are shared open metadata schemas, core ontologies, public ontology services. Several practical applications testing demonstrating usefulness overviewed fields eculture, ehealth, egovernment, elearning, ecommerce.
We introduce interactive intent modeling, where the user directs exploratory search by providing feedback for estimates of intents. The estimated intents are visualized interaction on an Intent Radar, a novel visual interface that organizes onto radial layout relevant close to center visualization and similar have angles. can give intents, from which system learns visualizes improved estimates. systematically evaluated effect modeling in mixed-method task-based information seeking setting...
This paper presents the CultureSampo system from viewpoint of publishing heterogeneous linked data as a service. Discussed are problems converting legacy into data, well challenge making massively
Term-Relevance Prediction from Brain Signals (TRPB) is proposed to automatically detect relevance of text information directly brain signals. An experiment with forty participants was conducted record neural activity while providing judgments stimuli for a given topic. High-precision scientific equipment used quantify across 32 electroencephalography (EEG) channels. A classifier based on multi-view EEG feature representation showed improvement up 17% in prediction signals alone. Relevance...
The concept of temporal visual attention in dynamic contents, such as videos, has been much less studied than its spatial counterpart, i.e., salience. Yet, is useful for many downstream tasks, video compression and summarisation, or monitoring users' engagement with information. Previous work considered quantifying a salience score from spatio-temporal user agreements gaze data. Instead gaze-based content-based approaches, we explore to what extent only brain signals can reveal attention. We...
Abstract Since the recent emergence of electronic literature resources, researchers have begun to adopt new information‐seeking practices. The purpose this research is investigate information needs and searching behaviors researchers, their implications for search tools. We conducted mixed‐method case studies involving interviews, diary logs, observations computer scientists followed by a web‐based survey validate our findings. results show that science following main purposes seeking...
Exploratory search confront users with challenges in expressing intents as the current interfaces require investigating result listings to identify directions, iterative typing, and reformulating queries. We present design of Exploration Wall, a touch-based user interface that allows incremental exploration sense-making large information spaces by combining entity search, flexible use entities query parameters, spatial configuration streams are visualized for interaction. Entities can be...
Exploratory search requires the system to assist user in comprehending information space and expressing evolving intents for iterative exploration retrieval of information. We introduce interactive intent modeling, a technique that models user’s visualizes them as keywords interaction. The can provide feedback on keywords, from which learns an improved estimate retrieves report experiments comparing variants implementing modeling control system. Data comprising logs, interaction essay...
Information retrieval systems often consider search-session and immediately preceding web-browsing history as the context for predicting users’ present information needs. However, such is only available when a user’s needs originate from web or users have issued queries in search session. Here, we study effect of more extensive recorded everyday digital activities by monitoring all interacted with communicated using personal computers. Twenty individuals were recruited 14 days 24/7...
Fairness and relevance are two important aspects of recommender systems (RSs). Typically, they evaluated either (i) separately by individual measures fairness relevance, or (ii) jointly using a single measure that accounts for with respect to relevance. However, approach often does not provide reliable joint estimate the goodness models, as it has different best models: one another Approach is also problematic because these tend be ad-hoc do relate well traditional measures, like NDCG....
Language reconstruction from non-invasive brain recordings has been a long-standing challenge. Existing research addressed this challenge with classification setup, where set of language candidates are pre-constructed and then matched the representation decoded recordings. Here, we propose method that addresses through auto-regressive generation, which directly uses functional magnetic resonance imaging (fMRI) as input for large model (LLM), mitigating need candidates. While an LLM can...
Collaborative idea generation leverages social interactions and knowledge sharing to spark diverse associations produce creative ideas. Information exploration systems expand the current context by suggesting novel but related concepts. In this paper we introduce InspirationWall, an unobtrusive display that speech recognition information enhance ongoing session with automatically retrieved concepts relate conversation. We evaluated system in six sessions of 20 minutes small groups two...
We introduce topic-relevance map, an interactive search result visualization that assists rapid information comprehension across a large ranked set of results. The map visualizes topical overview the space as keywords with respect to two essential retrieval measures: relevance and similarity. Non-linear dimensionality reduction is used embed high-dimensional keyword representations data into angles on radial layout. Relevance estimated by ranking method visualized radiuses As result, similar...
We present visual re-ranking, an interactive visualization technique for multi-aspect information retrieval. In search, the need of user consists more than one aspect or query simultaneously. While and search interface techniques improving interpretation results have been proposed, current research lacks understanding on how useful these are user: whether they lead to quantifiable benefits in perceiving result space allow faster, precise Our visualizes relevance document density a...
Previous research investigated how to leverage the new type of social data available on web, e.g., tags, ratings and reviews, in recommending personalizing information. However, previous works mainly focused predicting using collaborative filtering or quantifying personalized ranking quality simulations. As a consequence, effect information user's search information-selection behavior remains elusive. The objective our is investigate effects users' interactive behavior. We present...