Johan Hagelbäck
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Johan Hagelbäck has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Petar Jerčić
Blekinge Institute of Technology
2 shared publications - Anders Ardö
Department of Electrical and Information Technology , Lund University , Lund , Sweden
1 shared publication - Craig A. Lindley
CSIRO
1 shared publication - Koraljka Golub
Linnaeus University, Kalmar, Sweden
1 shared publication - Stefan J. Johansson
School of Computing, Blekinge Institute of Technology, Karlskrona, Sweden
1 shared publication - Veronica Sundstedt
Department of Computer Science, Blekinge Institute of Technology, Karlskrona, Sweden
1 shared publication
Latest Publications
- Open Access
- 1 February 2020
- Published by Walter de Gruyter GmbH in Journal of Data and Information Science
- Vol. 5 (1) , 18-38
Automatic Classification of Swedish Metadata Using Dewey Decimal Classification: A Comparison of Approaches
Purpose: With more and more digital collections of various information resources becoming available, also increasing is the challenge of assigning subject index terms and classes from quality knowledge organization systems. While the ultimate purpose is to understand the value of automatically produced Dewey Decimal Classification (DDC) classes for Swedish digital collections, the paper aims to evaluate the performance of six machine learning algorithms as well as a string-matching algorithm based on characteristics of DDC. Design/methodology/approach: State-of-the-art machine learning algorithms require at least 1,000 training examples per class. The complete data set at the time of research involved 143,838 records which had to be reduced to top three hierarchical levels of DDC in order to provide sufficient training data (totaling 802 classes in the training and testing sample, out of 14,413 classes at all levels). Findings: Evaluation shows that Support Vector Machine with linear kernel outperforms other machine learning algorithms as well as the string-matching algorithm on average; the string-matching algorithm outperforms machine learning for specific classes when characteristics of DDC are most suitable for the task. Word embeddings combined with different types of neural networks (simple linear network, standard neural network, 1D convolutional neural network, and recurrent neural network) produced worse results than Support Vector Machine, but reach close results, with the benefit of a smaller representation size. Impact of features in machine learning shows that using keywords or combining titles and keywords gives better results than using only titles as input. Stemming only marginally improves the results. Removed stop-words reduced accuracy in most cases, while removing less frequent words increased it marginally. The greatest impact is produced by the number of training examples: 81.90% accuracy on the training set is achieved when at least 1,000 records per class are available in the training set, and 66.13% when too few records (often less than 100 per class) on which to train are available—and these hold only for top 3 hierarchical levels (803 instead of 14,413 classes). Research limitations: Having to reduce the number of hierarchical levels to top three levels of DDC because of the lack of training data for all classes, skews the results so that they work in experimental conditions but barely for end users in operational retrieval systems. Practical implications: In conclusion, for operative information retrieval systems applying purely automatic DDC does not work, either using machine learning (because of the lack of training data for the large number of DDC classes) or using string-matching algorithm (because DDC characteristics perform well for automatic classification only in a small number of classes). Over time, more training examples may become available, and DDC may be enriched with synonyms in order to enhance accuracy of automatic classification which may also benefit information retrieval performance based on DDC. In order for quality information services to reach the objective of highest possible precision and recall, automatic classification should never be implemented on its own; instead, machine-aided indexing that combines the efficiency of automatic suggestions with quality of human decisions at the final stage should be the way for the future. Originality/value: The study explored machine learning on a large classification system of over 14,000 classes which is used in operational information retrieval systems. Due to lack of sufficient training data across the entire set of classes, an approach complementing machine learning, that of string matching, was applied. This combination should be explored further since it provides the potential for real-life applications with large target classification systems.
- 1 February 2020
- Published by Walter de Gruyter GmbH in Journal of Data and Information Science
- Vol. 5 (1) , 18-38
Open AccessAutomatic Classification of Swedish Metadata Using Dewey Decimal Classification: A Comparison of Approaches
Purpose: With more and more digital collections of various information resources becoming available, also increasing is the challenge of assigning subject index terms and classes from quality knowledge organization systems. While the ultimate purpose is to understand the value of automatically produced Dewey Decimal Classification (DDC) classes for Swedish digital collections, the paper aims to evaluate the performance of six machine learning algorithms as well as a string-matching algorithm based on characteristics of DDC. Design/methodology/approach: State-of-the-art machine learning algorithms require at least 1,000 training examples per class. The complete data set at the time of research involved 143,838 records which had to be reduced to top three hierarchical levels of DDC in order to provide sufficient training data (totaling 802 classes in the training and testing sample, out of 14,413 classes at all levels). Findings: Evaluation shows that Support Vector Machine with linear kernel outperforms other machine learning algorithms as well as the string-matching algorithm on average; the string-matching algorithm outperforms machine learning for specific classes when characteristics of DDC are most suitable for the task. Word embeddings combined with different types of neural networks (simple linear network, standard neural network, 1D convolutional neural network, and recurrent neural network) produced worse results than Support Vector Machine, but reach close results, with the benefit of a smaller representation size. Impact of features in machine learning shows that using keywords or combining titles and keywords gives better results than using only titles as input. Stemming only marginally improves the results. Removed stop-words reduced accuracy in most cases, while removing less frequent words increased it marginally. The greatest impact is produced by the number of training examples: 81.90% accuracy on the training set is achieved when at least 1,000 records per class are available in the training set, and 66.13% when too few records (often less than 100 per class) on which to train are available—and these hold only for top 3 hierarchical levels (803 instead of 14,413 classes). Research limitations: Having to reduce the number of hierarchical levels to top three levels of DDC because of the lack of training data for all classes, skews the results so that they work in experimental conditions but barely for end users in operational retrieval systems. Practical implications: In conclusion, for operative information retrieval systems applying purely automatic DDC does not work, either using machine learning (because of the lack of training data for the large number of DDC classes) or using string-matching algorithm (because DDC characteristics perform well for automatic classification only in a small number of classes). Over time, more training examples may become available, and DDC may be enriched with synonyms in order to enhance accuracy of automatic classification which may also benefit information retrieval performance based on DDC. In order for quality information services to reach the objective of highest possible precision and recall, automatic classification should never be implemented on its own; instead, machine-aided indexing that combines the efficiency of automatic suggestions with quality of human decisions at the final stage should be the way for the future. Originality/value: The study explored machine learning on a large classification system of over 14,000 classes which is used in operational information retrieval systems. Due to lack of sufficient training data across the entire set of classes, an approach complementing machine learning, that of string matching, was applied. This combination should be explored further since it provides the potential for real-life applications with large target classification systems.
- 25 August 2018
- Published by Springer Nature
- p. 127-138
Physiological Affect and Performance in a Collaborative Serious Game Between Humans and an Autonomous Robot
- 25 August 2018
- Published by Springer Nature
- p. 127-138
Physiological Affect and Performance in a Collaborative Serious Game Between Humans and an Autonomous Robot
- Open Access
- 25 October 2017
- Published by Springer Nature in International Journal of Social Robotics
- Vol. 10 (1) , 115-129
The Effect of Emotions and Social Behavior on Performance in a Collaborative Serious Game Between Humans and Autonomous Robots
- 25 October 2017
- Published by Springer Nature in International Journal of Social Robotics
- Vol. 10 (1) , 115-129
Open AccessThe Effect of Emotions and Social Behavior on Performance in a Collaborative Serious Game Between Humans and Autonomous Robots
- Open Access
- 1 August 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 23254270,p. 139-145
A study on human like characteristics in real time strategy games
Computer controlled characters (NPCs) are important in any video game to make the game world interesting, give more depth to a game and make the game playable. In almost any game the player has to cooperate with, fight against or interact with NPCs. This is especially true for single-player games but NPCs are also important in most multi-player games. When creating NPCs the developers often strive to create human like characters that behave reasonably intelligent in most cases. We have performed a study aiming to give an idea of the characteristics of human like NPCs in real-time strategy (RTS) games. In the study participants were asked to watch a recording of an RTS game and decide and motivate if the players in the game were controlled by a human player or a computer. We recorded matches were human players played against bots as well as bots playing against other bots. The results were categorized into different groups and they showed that some characteristics, for example simultaneous movement, are perceived as very bot-like and other things such as ability to try different tactics are perceived as humanlike.
- 1 August 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 23254270,p. 139-145
Open AccessA study on human like characteristics in real time strategy games
Computer controlled characters (NPCs) are important in any video game to make the game world interesting, give more depth to a game and make the game playable. In almost any game the player has to cooperate with, fight against or interact with NPCs. This is especially true for single-player games but NPCs are also important in most multi-player games. When creating NPCs the developers often strive to create human like characters that behave reasonably intelligent in most cases. We have performed a study aiming to give an idea of the characteristics of human like NPCs in real-time strategy (RTS) games. In the study participants were asked to watch a recording of an RTS game and decide and motivate if the players in the game were controlled by a human player or a computer. We recorded matches were human players played against bots as well as bots playing against other bots. The results were categorized into different groups and they showed that some characteristics, for example simultaneous movement, are perceived as very bot-like and other things such as ability to try different tactics are perceived as humanlike.