Björn Gambäck

Björn Gambäck
Norwegian University of Science and Technology | NTNU · Department of Computer and Information Science

PhD (tekn. dr.) Computer and System Sciences

About

112
Publications
32,967
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,528
Citations
Citations since 2017
30 Research Items
1084 Citations
2017201820192020202120222023050100150200250
2017201820192020202120222023050100150200250
2017201820192020202120222023050100150200250
2017201820192020202120222023050100150200250

Publications

Publications (112)
Chapter
Previous work on music generation and transformation has commonly targeted single instrument or single melody music. Here, in contrast, five music genres are used with the goal to achieve selective remixing by using domain transfer methods on spectrogram images of music. A pipeline architecture comprised of two independent generative adversarial ne...
Chapter
In contrast to most recent models that generate an entire image at once, the paper introduces a new architecture for generating images one pixel at a time using a Compositional Pattern-Producing Network (CPPN) as the generator part in a Generative Adversarial Network (GAN), allowing for effective generation of visually interesting images with artis...
Preprint
Full-text available
In this paper, we present the results of the SemEval-2020 Task 9 on Sentiment Analysis of Code-Mixed Tweets (SentiMix 2020). We also release and describe our Hinglish (Hindi-English) and Spanglish (Spanish-English) corpora annotated with word-level language identification and sentence-level sentiment labels. These corpora are comprised of 20K and 1...
Preprint
Full-text available
Information on social media comprises of various modalities such as textual, visual and audio. NLP and Computer Vision communities often leverage only one prominent modality in isolation to study social media. However, the computational processing of Internet memes needs a hybrid approach. The growing ubiquity of Internet memes on social media plat...
Article
Sentiment analysis is a circumstantial analysis of text, identifying the social sentiment to better understand the source material. The article addresses sentiment analysis of an English-Hindi and English-Bengali code-mixed textual corpus collected from social media. Code-mixing is an amalgamation of multiple languages, which previously mainly was...
Chapter
The paper tackles the question of evolvable media repositories, i.e., local pools of media files that are retrieved over the Internet and that are ever-renovated with new, related files in an evolutionary fashion. The herein proposed method encodes genotypic space by virtue of simple undirected graphs of natural language tokens that represent web q...
Article
"Community" in social networks is a nebulous concept. A community is generally assumed to be formed by people who possess similar attributes or characteristics, also known as "homophily". Although there has been a lot of research on community detection based on network topology, the semantic interpretation of communities is rarely studied. The pres...
Article
Full-text available
A community in social networks is generally assumed to be composed of a group of individuals with similar characteristics. Although there has been a plethora of work on understanding network topologies (edge density, clustering coefficient, etc.) within an online community, the psycho-sociological compositions of social network communities have har...
Conference Paper
Full-text available
The paper describes a metacreative system for real time algorithmic composition of audio mashups and synthetic soundscapes that pivots on evolvable media repositories, i.e., local pools of related media content that are retrieved, and ever-renovated, over the WWW in an evolutionary fashion. The model also involves a sophisticated soundscape generat...
Article
This article addresses language identification at the word level in Indian social media corpora taken from Facebook, Twitter and WhatsApp posts that exhibit code-mixing between English-Hindi, English-Bengali, as well as a blend of both language pairs. Code-mixing is a fusion of multiple languages previously mainly associated with spoken language, b...
Chapter
In order to improve the performance of a deep-learning neural network, the paper outlines a stack-based approach incorporating various information sources. A named entity recognition system for Amharic was implemented using a recurrent neural network, a bi-directional long short term memory model. Word vectors based on semantic information were bui...
Chapter
Identifying sarcasm in text is a challenging task which can be difficult also for humans, in particular in very short texts with little explicit context, such as tweets (Twitter messages). The paper presents a comparison of three sets of tweets marked for sarcasm, two annotated manually and one annotated using the common strategy of relying on the...
Conference Paper
Full-text available
Domain Adaptation in Machine Translation means to take a machine translation system that is restricted to work in a specific context and to enable the system to translate text from a different domain. The paper presents a two-step domain adaptation strategy, by first making use of unlabeled training material through an unsupervised algorithm, the S...
Research
The paper reports work on collecting and annotating code-mixed English-Hindi social media text (Twitter and Facebook messages), and experiments on automatic tagging of these corpora, using both a coarse-grained and a fine-grained part-of-speech tag set. We compare the performance of a combination of language specific taggers to that of applying fou...
Conference Paper
Full-text available
The paper describes the first sophisticated negation scope detection system for Twitter sentiment analysis. The system has been evaluated both on existing corpora from other domains and on a corpus of English Twitter data (tweets) annotated for negation. It produces better results than what has been reported in other domains and improves the perfor...
Conference Paper
Full-text available
Agent-based models of language evolution have received a lot of attention in the last two decades. Researchers wish to understand the origin of language, and aim to compensate for the lacking empirical evidence by utilizing methods from computer science and artificial life. The paper looks at the main theories of language evolution: biological evol...
Conference Paper
Full-text available
Most distributional models of word similarity represent a word type by a single vector of contextual features, even though, words commonly have more than one sense. The multiple senses can be captured by employing several vectors per word in a multi-prototype distributional model, prototypes that can be obtained by first constructing all the contex...
Conference Paper
A closer look at how users perform search is needed in order to best design a more efficient next generation sentiment search engine and understand fundamental behaviours involved in online review/opinion search processes. The paper proposes utilizing personalized search, eye tracking and sentiment analysis for better understanding of end-user beha...
Article
The "How Was Your Day" (HWYD) companion is an embodied conversational agent that can discuss work-related issues, entering free-form dialogues while discussing issues surrounding a typical work day. The open-ended nature of these interactions requires new models of evaluation. Here, we describe a paradigm and methodology for evaluating the main asp...
Article
Arguably, the most important difference between machines and humans is that humans have feelings. For several decades researchers have been trying to create methods to simulate sentimentality for machines, and currently Sentiment Analysis is the hottest, most demanding, and rapidly growing task in the language processing field. Sentiment analysis o...
Conference Paper
A system performing sentiment analysis needs some prior knowledge which can be acquired, for example, by manual annotation only or by expanding a human-annotated sentiment lexicon by automatic means. However, users are seldom interested in all sentimental aspects at once, but rather look for opinion changes of some person ("Who") during some time p...
Article
Full-text available
The paper discusses several complex transfer problems and their prospective solutions within an English-to-Swedish spoken language translation system. The emphasis in the text is on transfer problems which are not lexically triggered, concentrating mainly on the translation of differences in mood and tense. Laying the groundworks for the trans-lati...
Conference Paper
Full-text available
Word Translation Disambiguation is the task of selecting the best translation(s) for a source word in a certain context, given a set of translation candidates. Most approaches to this problem rely on large word-aligned parallel corpora, resources that are scarce and expensive to build. In contrast, the method presented in this paper requires only l...
Conference Paper
Full-text available
Current sentiment analysis systems rely on static (context independent) sentiment lexica with proximity based fixed-point prior polarities. However, sentiment-orientation changes with context and these lexical resources give no indication of which value to pick at what context. The general trend is to pick the highest one, but which that is may var...
Article
Full-text available
In this paper we address the Sentiment Analysis problem from the end user's perspective. An end user might desire an automated at-a-glance presentation of the main points made in a single review or how opinion changes time to time over multiple documents. To meet the requirement we propose a relatively generic opinion 5Ws structurization, further u...
Conference Paper
Full-text available
In this paper we address the Sentiment Analysis problem from the end user's perspective. An end user might desire an automated at-a-glance presentation of the main points made in a single review or how opinion changes time to time over multiple documents. To meet the requirement we propose a relatively generic opinion 5Ws structurization, further u...
Article
Multimodal conversational spoken dialogues using physical and virtual agents provide a potential interface to motivate and support users in the domain of health and fitness. This paper describes how such multimodal conversational Companions can be implemented to support their owners in various pervasive and mobile settings. We present concrete syst...
Conference Paper
Full-text available
Active learning techniques were employed for classification of dialogue acts over two dialogue corpora, the English human-human Switchboard corpus and the Spanish human-machine Dihana corpus. It is shown clearly that active learning improves on a baseline obtained through a passive learning approach to tagging the same data sets. An error reduction...
Article
Full-text available
A major bottleneck for promoting use of computers and the Internet is that many languages lack access to basic tools that would make it possible for people to access ICT in their own language. The paper describes the development a set of such resources for the processing of Amharic, the working language of the Ethiopian government. The primary goal...
Book
The current state of dialogue technology has come a long way since its beginning in the 1950s: dialogue technology now provides interactive service agents, while research explores various aspects of multimodal and multiparty communication so as to improve natural and social aspects of dialogue systems. In this workshop, interest is focussed especia...
Article
Full-text available
We present work aimed at compiling an Amharic corpus from the Web and automatically categorizing the texts. Amharic is the second most spoken Semitic language in the World (after Arabic) and used for countrywide communication in Ethiopia. It is highly inflectional and quite dialectally diversified. We discuss the issues of compiling and annotating...
Conference Paper
Full-text available
Multimodal conversational spoken dia- logues using physical and virtual agents provide a potential interface to motivate and support users in the domain of health and fitness. The paper presents a multi- modal conversational Companion system focused on health and fitness, which has both a stationary and a mobile component.
Conference Paper
Full-text available
The paper presents an evaluation of four clustering algo- rithms: k-means, average linkage, complete linkage, and Ward's method, with the latter three being different hier- archical methods. The quality of the clusters created by the algorithms was measured in terms of cluster cohesive- ness and semantic cohesiveness, and both quantitative and pred...
Article
Full-text available
The paper describes a tagged corpus of Amharic news texts and some machine learning-based tagging experiments that have been carried out on the corpus. Amharic is the second most spoken Semitic language in the World (after Arabic) and used for countrywide communication in Ethiopia. It is highly inflectional and quite dialectally diversified. We hav...
Article
Full-text available
The paper describes a set of experiments involving the application of three state-of- the-art part-of-speech taggers to Ethiopian Amharic, using three different tagsets. The taggers showed worse performance than previously reported results for Eng- lish, in particular having problems with unknown words. The best results were obtained using a Maximu...
Article
Full-text available
In this paper we present the EU-IST project DUMAS (Dynamic Universal Mobility for Adaptive Speech Interfaces), and discuss adaptation and robust information processing as realized in AthosMail, a speech-based multilingual email application developed within the project. One of the goals of the research conducted in the project has been to develop a...
Article
Full-text available
Multimodal conversational spoken dialogues using physical and virtual agents provide a potential interface to motivate and support users in the domain of health and fitness. In this paper we present how such multimodal conversational Companions can be implemented to support their owners in various pervasive and mobile settings. In particular, we fo...
Article
Full-text available
The paper introduces a Mobile Companion prototype, which helps users to plan and keep track of their exercise activities via an interface based mainly on speech input and output. The Mobile Companion runs on a PDA and is based on a stand-alone, speaker-independent solution, making it fairly unique among mobile spoken dialogue systems, where the com...
Conference Paper
Full-text available
The paper discusses an Amharic speaker independent contin- uous speech recognizer based on an HMM/ANN hybrid ap- proach. The model was constructed at a context dependent phone part sub-word level with the help of the CSLU Toolkit. A promising result of 74.28% word and 39.70% sentence recog- nition rate was achieved. These are the best figures repor...
Article
Full-text available
AthosMail is a multilingual spoken dialogue system for reading of e-mail messages. The key features of the application are adaptivity and the integration of different approaches for spoken interaction. The application has flexible system structure supporting multiple components for both different and same purposes. The AthosMail system includes com...
Article
In this paper we present the EU-IST project DUMAS (Dynamic Universal Mobility for Adaptive Speech Interfaces), and discuss adaptation and robust information processing as realized in AthosMail, a speechbased multilingual email application developed within the project. One of the goals of the research conducted in the project has been to develop a s...
Article
Full-text available
Gaming is a highly relevant application area for Intelligent Agents and Human Computer Interaction (HCI). Computer games bring us a full set of new gaming experiences where synthetic characters take on the main role. Using affective input in the interaction with a game and in particular with a character is a recent and fairly unexplored dimension....
Article
Full-text available
The DUMAS project develops speech-based applications that are adaptable to different users and domains. The paper describes the project's robust semantic analysis strategy, used both in the generic framework for the development of multilingual speech-based dialogue systems which is the main project goal, and in the initial test application, a mobil...
Article
Full-text available
When developing adaptive speech-based multilingual interaction systems, we need representative data on the user's behaviour. In this paper we focus on a data collection method pertaining to adaptation in the user's interaction with the system. We describe a multi-session group scenario for Wizard of Oz studies with two novel features: firstly, inst...
Article
Full-text available
Gaming is a highly relevant application area for Intelligent Agents and Human Computer Interaction (HCI). Computer games bring us a full set of new gaming experiences where synthetic characters take on the main role. Using affective input in the interaction with a game and in particular with a character is a recent and fairly unexplored dimension....
Article
The paper discusses how compositional semantics is implemented in the Verbmobil speech-to-speech translation system using LUD, a description language for underspecified discourse representation structures. The description language and its formal interpretation in DRT are described as well as its implementation together with the architecture of the...
Article
Full-text available
The DUMAS project constructs a generic framework for the development of multilingual speech-based dia-logues systems. As an initial test of the generic frame-work we will build a mobile phone-based e-mail in-terface whose functionality can be adapted to differ-ent users, different situations and tasks. The paper describes the semantic processing wh...
Article
It is only during the last few years that attention has started to shift from pure textbased retrieval towards other media. Information retrieval from spoken documents is analogous to text-based retrieval; however, accessing audio documents causes some extra problems, in particular with respect to document segmentation, choice of indexing features,...
Conference Paper
The implementation of a unification-based lexicon is dis- cussed as well as the morphological rules needed for mapping between the lexicon and grammar. It is shown how different feature usages can be utilized in the implementation to reach the intended surface word-form matches, with the correct feature settings. A novelty is the way features are u...
Article
Full-text available
The paper discusses the lessons we have learned from the work on building a reusable toolset for Swedish within the framework of GATE, the General Architecture for Text Engineering, from the University of Sheffield, UK. We describe our toolbox svensk and the reasons behind the choices made in the design, as well as the overall conclusions for langu...
Article
Traditionally, the level of reusability of language processing resources within the research community has been very low. Most of the recycling of linguistic resources has been concerned with reuse of data, e.g., corpora, lexica, and grammars, while the algorithmic resources far too seldom have been shared between different projects and institution...
Article
Full-text available
Often, information retrieval from various other media is analogous to text-based retrieval; however, accessing documents in e.g. audio or video formats causes some extra problems, in particular with respect to document segmentation, choice of indexing features, and robustness. We review these difficulties, together with some previous attempts to ov...
Article
The svensk project has developed an integrated toolbox of language processing components and resources for Swedish. Priority has been given to developing the toolbox for teaching (such as student projects) and research projects where a natural language interface is required. The toolbox is based on the GATE language engineering platform and will in...
Article
Full-text available
iii Abstract This booklet introduces natural-language processing in general and the way it is presently carried out at SICS. The overall goal of any system for naturallanguage processing system is to translate an input utterance stated in a natural language (such as English or Swedish) to some type of computer internal representation. Doing this re...