Abstract | ||
---|---|---|
This paper presents a novel discriminative learning technique for label sequences based on a combination of the two most success- ful learning algorithms, Support Vector Ma- chines and Hidden Markov Models which we call Hidden Markov Support Vector Ma- chine. The proposed architecture handles dependencies between neighboring labels us- ing Viterbi decoding. In contrast to stan- dard HMM training, the learning procedure is discriminative and is based on a maxi- mum/soft margin criterion. Compared to previous methods like Conditional Random Fields, Maximum Entropy Markov Models and label sequence boosting, HM-SVMs have a number of advantages. Most notably, it is possible to learn non-linear discriminant functions via kernel functions. At the same time, HM-SVMs share the key advantages with other discriminative methods, in partic- ular the capability to deal with overlapping features. We report experimental evaluations on two tasks, named entity recognition and part-of-speech tagging, that demonstrate the competitiveness of the proposed approach. |
Year | Venue | Keywords |
---|---|---|
2003 | ICML | conditional random field,kernel function,support vector,discrimination learning,markov model,hidden markov model,support vector machine,maximum entropy,viterbi decoder |
Field | DocType | Citations |
Conditional random field,Maximum-entropy Markov model,Sequence labeling,Pattern recognition,Forward algorithm,Markov model,Computer science,Variable-order Markov model,Artificial intelligence,Hidden Markov model,Viterbi algorithm,Machine learning | Conference | 257 |
PageRank | References | Authors |
29.54 | 9 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
yasemin altun | 1 | 2463 | 150.46 |
Ioannis Tsochantaridis | 2 | 2861 | 155.43 |
Thomas Hofmann | 3 | 10064 | 1001.83 |
t fawcett n mishra | 4 | 257 | 29.54 |