Machine learning investigation of machine iearning and there application to WSD
By: Patil, Harshada Suresh
.
Contributor(s): Wadhai, Sheetal A.
Publisher: Haryana IOSR - International Organization of Scientific Research 2022Edition: Vol.24(3), May-Jun.Description: 26-31p.Subject(s): Computer EngineeringOnline resources: Click here
In:
IOSR Journal of Computer Engineering (IOSR-JCE)Summary: Word Sense Disambiguation (WSD) is a task in Natural Language Processing (NLP) that determines the right
meaning (sense) for a given word in a text or speech that is distinguished from alternative meanings. senses that
could be attributed to the term These senses could be considered the target. A classification problem's labels
That is, Machine Learning (ML) appears to be a viable option. A solution to this problem This paper
investigates the potential applications of the methods and techniques of Machine Learning was used to handle
the WSD problem. The first issue addressed was the adaption of various ML algorithms to deal with word
senses serving as classifications Following that, a comparison of various methodologies is carried out under the
the same circumstances The conventional precision and recall measurements, as well as agreement rates and
kappa statistics, are used to compare the results of these approaches.The second topic investigated is the cross-
corpora use of supervised Machine Learning.WSD learning systems to assess generalisation capacity across
corpora and domains.The results found are quite unsatisfactory, calling into serious question the possibility of
creating a sufficiently broad training corpus (labelled or unlabeled), and the manner in which it is used to
create a general-purpose Word Sense Tagger, samples should be used. The application of the use of unlabeled
data to train classifiers for Word Sense Disambiguationis a relatively common practise. Difficult line of
research in order to create a truly robust, full, and accurate. Tagger for Word Sense As a result of this, the
following topic The application is the subject of this paper. Considering two WSD bootstrapping methods:
Transductive Support Vector Machines and Steven Abney's Greedy Agreement bootstrapping algorithm.
| Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
|---|---|---|---|---|---|---|
Articles Abstract Database
|
School of Engineering & Technology Archieval Section | Not for loan | 2022-2112 |
Word Sense Disambiguation (WSD) is a task in Natural Language Processing (NLP) that determines the right
meaning (sense) for a given word in a text or speech that is distinguished from alternative meanings. senses that
could be attributed to the term These senses could be considered the target. A classification problem's labels
That is, Machine Learning (ML) appears to be a viable option. A solution to this problem This paper
investigates the potential applications of the methods and techniques of Machine Learning was used to handle
the WSD problem. The first issue addressed was the adaption of various ML algorithms to deal with word
senses serving as classifications Following that, a comparison of various methodologies is carried out under the
the same circumstances The conventional precision and recall measurements, as well as agreement rates and
kappa statistics, are used to compare the results of these approaches.The second topic investigated is the cross-
corpora use of supervised Machine Learning.WSD learning systems to assess generalisation capacity across
corpora and domains.The results found are quite unsatisfactory, calling into serious question the possibility of
creating a sufficiently broad training corpus (labelled or unlabeled), and the manner in which it is used to
create a general-purpose Word Sense Tagger, samples should be used. The application of the use of unlabeled
data to train classifiers for Word Sense Disambiguationis a relatively common practise. Difficult line of
research in order to create a truly robust, full, and accurate. Tagger for Word Sense As a result of this, the
following topic The application is the subject of this paper. Considering two WSD bootstrapping methods:
Transductive Support Vector Machines and Steven Abney's Greedy Agreement bootstrapping algorithm.
Click on an image to view it in the image viewer
There are no comments for this item.