Epidemic spreading based word sense disambiguation of Hindi text

Jain, Goonjan; Lobiyal, D. K.

doi:10.1201/9780429444272-10

Cited by 3 publications

(5 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A technique to perform unsupervised WSD on a Hindi sentence using network agglomeration is proposed in [64]. We start by creating a graph G for the input sentence.…”

Section: Unsupervised Techniquesmentioning

confidence: 99%

A Literature Survey on Word Sense Disambiguation for the Hindi Language

Gujjar,

Mago,

Kumari

et al. 2023

Information

View full text Add to dashboard Cite

Word sense disambiguation (WSD) is a process used to determine the most appropriate meaning of a word in a given contextual framework, particularly when the word is ambiguous. While WSD has been extensively studied for English, it remains a challenging problem for resource-scarce languages such as Hindi. Therefore, it is crucial to address ambiguity in Hindi to effectively and efficiently utilize it on the web for various applications such as machine translation, information retrieval, etc. The rich linguistic structure of Hindi, characterized by complex morphological variations and syntactic nuances, presents unique challenges in accurately determining the intended sense of a word within a given context. This review paper presents an overview of different approaches employed to resolve the ambiguity of Hindi words, including supervised, unsupervised, and knowledge-based methods. Additionally, the paper discusses applications, identifies open problems, presents conclusions, and suggests future research directions.

show abstract

“…A technique to perform unsupervised WSD on a Hindi sentence using network agglomeration is proposed in [64]. We start by creating a graph G for the input sentence.…”

Section: Unsupervised Techniquesmentioning

confidence: 99%

A Literature Survey on Word Sense Disambiguation for the Hindi Language

Gujjar,

Mago,

Kumari

et al. 2023

Information

View full text Add to dashboard Cite

show abstract

“…Fuzzy Graph connectivity measures (FGCM) were proposed by Jain and Lobiyal (2015a, 2015b). It has been heavily used in wide applications of NLP for instance (Jain et al, 2019; Jain et al, 2022; Jain, Sharma, et al, 2021; Vij et al, 2017; Vij et al, 2019).…”

Section: Preliminariesmentioning

confidence: 99%

“…Fuzzy eigenvector centrality : According to this centrality, connections to nodes with high scores “add more to the score of the node in question than similar connections to nodes with low scores.” It comes in two flavours: Fuzzy PageRank and Fuzzy HITS (Jain & Lobiyal, 2015a, 2015b).…”

Section: Preliminariesmentioning

confidence: 99%

“…Hubs

h_{f} (v)

and authorities

a_{f} (v)

have been proposed by Jain and Lobiyal for fuzzy graphs (2015a, 2015b).

h_{f} (v) = \sum_{(u, v) \in E} μ_{italicuv} a_{f} (u)

a_{f} (v) = \sum_{(u, v) \in E} μ_{italicuv} h_{f} (u)

Where

μ_{italicuv}

is the strength of the edge connecting vertices u and v .…”

Section: Preliminariesmentioning

confidence: 99%

“…Fuzzy

{betweennesscent}_{f} (v)

is a fraction of fuzzy betweenness ( v ) to the largest pair of vertices that do not have v for a vertex in a fuzzy graph (Jain & Lobiyal, 2015a, 2015b).

{Fuzzy betweennesscent}_{f} (v) = \frac{{betweenness}_{f} (v)}{(|V| - 1) (|V| - 2)}

…”

Section: Preliminariesmentioning

confidence: 99%

See 2 more Smart Citations

Code‐mixed Hindi‐English text correction using fuzzy graph and word embedding

2023

View full text Add to dashboard Cite

Interaction via social media involves frequent code‐mixed text, spelling errors and noisy elements, which creates a bottleneck in the performance of natural language processing applications. This proposed work is the first approach for code‐mixed Hindi‐English social media text that comprises language identification, detection and correction of non‐word (Out of Vocabulary) errors as well as real‐word errors occurring simultaneously. Each identified language (Devanagari Hindi, Roman Hindi, and English) has its own complexities and challenges. Errors are detected individually for each language and a suggestive list of the erroneous words is created. After this, a fuzzy graph between different words of the suggestive lists is generated using various semantic relations in Hindi WordNet. Word embeddings and Fuzzy graph‐based centrality measures are used to find the correct word. Several experiments are performed on different social media datasets taken from Instagram, Twitter, YouTube comments, Blogs, and WhatsApp. The experimental results demonstrate that the proposed system corrects out‐of‐vocabulary words as well as real‐word errors with a maximum recall of 0.90 and 0.67, respectively for Dev_Hindi and 0.87 and 0.66, respectively for Rom_Hindi. The proposed method is also applied for state‐of‐art sentiment analysis approaches where the F1‐score has been visibly improved.

show abstract