J. Kangas - Academia.edu (original) (raw)
Uploads
Papers by J. Kangas
Series in Machine Perception and Artificial Intelligence, 1999
Perspectives in Neural Computing, 1998
Ponderomotive interaction between ion cyclotron waves and heavy ions in the magnetosphere influen... more Ponderomotive interaction between ion cyclotron waves and heavy ions in the magnetosphere influences deeply the particle and wave behaviour, especially in the near-equatorial regions. On the one hand, presence of heavy ions gives rise to the standing ion cyclotron resonator near the geomagnetic equator, which confines the ion cyclotron waves to a thin equatorial zone. On the other hand, ponderomotive forces induced by these standing waves redistribute heavy (light) ions forcing them to collect in the wave nodes (antinodes) of the electric field intensity. As a result, the spatial modulation in the chemical composition of plasma is developed. This in its turn changes structure and properties of the ion cyclotron resonator. Basic equations for solving the above mentioned self-consistent problem are given. The consequences, which are important in order to explain some unresolved questions of Pc1 observations, are discussed. The resonator is not an attribute of the solely Earth's m...
this paper, the self-organizing map was used to visualise thespeech signal and the speech signal ... more this paper, the self-organizing map was used to visualise thespeech signal and the speech signal variations in time, and from the mappingcreated by the SOM process it was further possible to construct some quantitativemeasures of the voice quality. In the studies the voice quality of vowels wasanalyzed.In [15] another application of the self-organizing map algorithm forspeech signal analysis is described. In this paper the self-organized mapwas used to distinguish between the /s/ samples ...
Journal of Geophysical Research, 2001
Journal of Geophysical Research, 2002
Journal of Atmospheric and Solar-Terrestrial Physics, 2000
Journal of Atmospheric and Solar-Terrestrial Physics, 2007
This paper is devoted to the beautiful and mysterious Pc1 geoelectromagnetic waves that attract t... more This paper is devoted to the beautiful and mysterious Pc1 geoelectromagnetic waves that attract the attention of many researchers all over the world as an essential element of the space physics. It is concluded that in spite of the recent progress in the investigation of Pc1 waves, it is still necessary to make a careful study of some unsettled problems
This paper discusses a combination of two techniques for improving the recognition accuracy of on... more This paper discusses a combination of two techniques for improving the recognition accuracy of on-line handwritten character recognition: committee classification and adaptation to the user. A novel adaptive committee structure, namely the class-confidence critic combination (CCCC) scheme, is presented and evaluated. It is shown to be able to improve significantly on its member classifiers. Also the effect of having either more or less diverse sets of member classifiers is considered.
ABSTRACT Subsystems for online recognition of handwriting are needed in personal digital assistan... more ABSTRACT Subsystems for online recognition of handwriting are needed in personal digital assistants (PDA) and other portable handheld devices. We have developed a recognition system which enhances its accuracy by applying continuous adaptation to the user's writing style. The forms of adaptation we have experimented with take place simultaneously with the normal operation of the system and therefore, there is no need for separate training period of the device. The present implementation uses dynamic time warping (DTW) in matching the input characters with the stored prototypes. The DTW algorithm implemented with dynamic programming (DP) is, however both time and memory consuming. In our current research we have experimented with methods that transform the elastic templates to pixel images which can then be recognized by using statistical or neural classification. The particular neural classifier we have used is the local subspace classifier (LSC) of which we have developed an adaptive version
Series in Machine Perception and Artificial Intelligence, 1999
Perspectives in Neural Computing, 1998
Ponderomotive interaction between ion cyclotron waves and heavy ions in the magnetosphere influen... more Ponderomotive interaction between ion cyclotron waves and heavy ions in the magnetosphere influences deeply the particle and wave behaviour, especially in the near-equatorial regions. On the one hand, presence of heavy ions gives rise to the standing ion cyclotron resonator near the geomagnetic equator, which confines the ion cyclotron waves to a thin equatorial zone. On the other hand, ponderomotive forces induced by these standing waves redistribute heavy (light) ions forcing them to collect in the wave nodes (antinodes) of the electric field intensity. As a result, the spatial modulation in the chemical composition of plasma is developed. This in its turn changes structure and properties of the ion cyclotron resonator. Basic equations for solving the above mentioned self-consistent problem are given. The consequences, which are important in order to explain some unresolved questions of Pc1 observations, are discussed. The resonator is not an attribute of the solely Earth's m...
this paper, the self-organizing map was used to visualise thespeech signal and the speech signal ... more this paper, the self-organizing map was used to visualise thespeech signal and the speech signal variations in time, and from the mappingcreated by the SOM process it was further possible to construct some quantitativemeasures of the voice quality. In the studies the voice quality of vowels wasanalyzed.In [15] another application of the self-organizing map algorithm forspeech signal analysis is described. In this paper the self-organized mapwas used to distinguish between the /s/ samples ...
Journal of Geophysical Research, 2001
Journal of Geophysical Research, 2002
Journal of Atmospheric and Solar-Terrestrial Physics, 2000
Journal of Atmospheric and Solar-Terrestrial Physics, 2007
This paper is devoted to the beautiful and mysterious Pc1 geoelectromagnetic waves that attract t... more This paper is devoted to the beautiful and mysterious Pc1 geoelectromagnetic waves that attract the attention of many researchers all over the world as an essential element of the space physics. It is concluded that in spite of the recent progress in the investigation of Pc1 waves, it is still necessary to make a careful study of some unsettled problems
This paper discusses a combination of two techniques for improving the recognition accuracy of on... more This paper discusses a combination of two techniques for improving the recognition accuracy of on-line handwritten character recognition: committee classification and adaptation to the user. A novel adaptive committee structure, namely the class-confidence critic combination (CCCC) scheme, is presented and evaluated. It is shown to be able to improve significantly on its member classifiers. Also the effect of having either more or less diverse sets of member classifiers is considered.
ABSTRACT Subsystems for online recognition of handwriting are needed in personal digital assistan... more ABSTRACT Subsystems for online recognition of handwriting are needed in personal digital assistants (PDA) and other portable handheld devices. We have developed a recognition system which enhances its accuracy by applying continuous adaptation to the user's writing style. The forms of adaptation we have experimented with take place simultaneously with the normal operation of the system and therefore, there is no need for separate training period of the device. The present implementation uses dynamic time warping (DTW) in matching the input characters with the stored prototypes. The DTW algorithm implemented with dynamic programming (DP) is, however both time and memory consuming. In our current research we have experimented with methods that transform the elastic templates to pixel images which can then be recognized by using statistical or neural classification. The particular neural classifier we have used is the local subspace classifier (LSC) of which we have developed an adaptive version