Open Access
Influence of Size on Pattern-based Sequence Classification
M. van Zaanen,T. Gaustad,J. Feijen +2 more
- 01 Jan 2011
- pp 53-60
TL;DR: The number of classes in the classification task does not have an effect on the trends in the results, although, as could be expected, tasks with fewer classes typically lead to higher accuracies.
read more
Abstract: In this paper we investigate the impact of size of vocabulary, the number of classes in the classification task and the length of patterns in a pattern-based sequence classification approach. So far, the approach has been applied successfully to datasets classifying into two or four classes. We now show results on six different classification tasks ranging from five to fifty classes. In addition, classification results on three different encodings of the data, leading to different vocabulary sizes, are explored. The system in general clearly outperforms the baseline. The encodings with a larger vocabulary size outperform those with smaller vocabularies. When using fixed length patterns the results are better, but vary more compared to using a range of pattern lengths. The number of classes in the classification task does not have an effect on the trends in the results, although, as could be expected, tasks with fewer classes typically lead to higher accuracies.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
•Book
Information retrieval
C. J. Van Rijsbergen
- 01 Jan 1975
TL;DR: The major change in the second edition of this book is the addition of a new chapter on probabilistic retrieval, which I think is one of the most interesting and active areas of research in information retrieval.
822
•Proceedings Article
Ensemble based co-training
Jafar Tanha,M. van Someren,Hamideh Afsarmanesh +2 more
- 01 Jan 2011
TL;DR: This paper proposes a criterion for finding a subset of high-confidence predictions and error rate for a classifier in each iteration of the training process, and shows that the new method in almost all domains gives better results than the other methods.
•Proceedings Article
Learning interpretations using sequence classification
Menno van Zaanen,Janneke van de Loo +1 more
- 16 Aug 2012
TL;DR: A system that assigns interpretations, in the form of shallow semantic frame descriptions, to natural language sentences, by searching for relevant patterns, to identify the correct semantic frame and associated slot values is presented.
•Journal Article
Applying Pattern-based Classification to Sequences of Gestures
TL;DR: It is demonstrated that sequences of gestures contain additional information compared to gestures in isolation, which introduces PBSC as an effective method to incorporate time as an extra dimension in gestural communication, which can be extended to a wide range of sequential modalities.
1
References
•Book
Modern Information Retrieval
Ricardo Baeza-Yates,Berthier Ribeiro-Neto +1 more
- 15 May 1999
TL;DR: In this article, the authors present a rigorous and complete textbook for a first course on information retrieval from the computer science (as opposed to a user-centred) perspective, which provides an up-to-date student oriented treatment of the subject.
•Book
Information retrieval
C. J. Van Rijsbergen
- 01 Jan 1975
TL;DR: The major change in the second edition of this book is the addition of a new chapter on probabilistic retrieval, which I think is one of the most interesting and active areas of research in information retrieval.
822
•Proceedings Article
Classification of musical genre: a machine learning approach.
Roberto Basili,Alfredo Serafini,Armando Stellato +2 more
- 01 Jan 2004
TL;DR: This work investigates the impact of machine learning algorithms in the development of automatic music classification models aiming to capture genres distinctions by first creating a medium-sized collection of examples for widely recognized genres and then evaluating the performances of different learning algorithms.
•Proceedings Article
Online Database of Scores in the Humdrum File Format.
Craig Sapp
- 01 Jan 2005
TL;DR: KernScores, an online library of musical data currently consisting of over 5 million notes, has been created to assist projects dealing with the computational analysis of musical scores.