Improving Query Classification by Features’ Weight Learning

Improving Query Classification by Features’ Weight Learning

Show other versions (1)

This work is an attempt to enhance query classification in call routing applications. A new method has been introduced to learn weights from training data by means of a regression model. This work has investigated applying the tf-idf weighting method, but the approach is not limited to a specific me...

Full description

Bibliographic Details
Main Author: Abghari, Arash
Language:en
Published: 2013
Subjects:
Query Classification
Weight learning
Electrical and Computer Engineering
Online Access:http://hdl.handle.net/10012/7484
id ndltd-LACETR-oai-collectionscanada.gc.ca-OWTU.10012-7484
record_format oai_dc
spelling ndltd-LACETR-oai-collectionscanada.gc.ca-OWTU.10012-74842013-10-04T04:12:25ZAbghari, Arash2013-04-30T13:35:46Z2013-04-30T13:35:46Z2013-04-30T13:35:46Z2013http://hdl.handle.net/10012/7484This work is an attempt to enhance query classification in call routing applications. A new method has been introduced to learn weights from training data by means of a regression model. This work has investigated applying the tf-idf weighting method, but the approach is not limited to a specific method and can be used for any weighting scheme. Empirical evaluations with several classifiers including Support Vector Machines (SVM), Maximum Entropy, Naive Bayes, and k-Nearest Neighbor (k-NN) show substantial improvement in both macro and micro F1 measures.enQuery ClassificationWeight learningImproving Query Classification by Features’ Weight LearningThesis or DissertationElectrical and Computer EngineeringMaster of Applied ScienceElectrical and Computer Engineering
collection NDLTD
language en
sources NDLTD
topic Query Classification
Weight learning
Electrical and Computer Engineering
spellingShingle Query Classification
Weight learning
Electrical and Computer Engineering
Abghari, Arash
Improving Query Classification by Features’ Weight Learning
description This work is an attempt to enhance query classification in call routing applications. A new method has been introduced to learn weights from training data by means of a regression model. This work has investigated applying the tf-idf weighting method, but the approach is not limited to a specific method and can be used for any weighting scheme. Empirical evaluations with several classifiers including Support Vector Machines (SVM), Maximum Entropy, Naive Bayes, and k-Nearest Neighbor (k-NN) show substantial improvement in both macro and micro F1 measures.
author Abghari, Arash
author_facet Abghari, Arash
author_sort Abghari, Arash
title Improving Query Classification by Features’ Weight Learning
title_short Improving Query Classification by Features’ Weight Learning
title_full Improving Query Classification by Features’ Weight Learning
title_fullStr Improving Query Classification by Features’ Weight Learning
title_full_unstemmed Improving Query Classification by Features’ Weight Learning
title_sort improving query classification by features’ weight learning
publishDate 2013
url http://hdl.handle.net/10012/7484
work_keys_str_mv AT abghariarash improvingqueryclassificationbyfeaturesweightlearning
_version_ 1716601069946011648

Similar Items