M. Elius Ahmed

M. Elius Ahmed

Auto-generated profile, based on publicly available information

M. Elius Ahmed

Department of Computer Science and Engineering, Hajee Mohammad Danesh Science and Technology University, Dinajpur, Bangladesh

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 April 2021 in IEEE Access
Verified
Journal Article
Open access
ATAtabuzzaman
Atabuzzaman
SHShajalal
Shajalal
MAM. Elius Ahmed
M. Elius Ahmed
MAMasud Ibn Afjal
Masud Ibn Afjal
MAMasaki Aono
Masaki Aono
Semantic similarity between texts can be defined based on their meaning. Assessing the textual similarity is a prerequisite in almost all applications in the field of language processing and information retrieval. However, the diversity in the sentence structure makes it formidable to estimate the similarity. Some sentences pairs are lexicographically similar but semantically dissimilar. That is why the trivial lexical overlapping is not enough for measuring the similarity. To attain the semanticity of sentences, the context of the words and the structure of the sentence should be considered. In this paper, we propose a new method for capturing the semantic similarity between sentences based on their grammatical roles through word semantics. First, the sentences are divided grammatically into different parts where each part is considered as a grammatical role. Then multiple new measures are introduced to estimate the role-based similarity exploiting word semantics considering the sentence structure. The proposed similarity measures focus on inter-role and intra-role similarity between the sentence-pair. The word-level semantic information is extracted from a pre-trained word-embedding model. The performance of the proposed method was verified by conducting a wide range of experiments on the SemEval STS dataset. The experimental results indicated the effectiveness of the proposed method in terms of different standard evaluation metrics and outperformed some known related works.

19 December 2020
Verified
Conference Paper
Closed access
MAM. Elius Ahmed
M. Elius Ahmed
SHShajalal
Shajalal
ATAtabuzzaman
Atabuzzaman
MAMasaki Aono
Masaki Aono
Some texts can convey the same meaning without consisting of common words because of the distinct ramification of sentences. Thence traditional string overlapping is not enough for measuring the semantic similarity. For the acquirement of semantic textual similarity, the word semantics and their senses should be considered. In this paper, we proposed a ratio considering words overlapping. The ratio is calculated in terms of common words and the uncommon words of two texts. We employed WordNet components to estimate semantic similarity. Various existing WordNet measures emphasize different relationships and information of words in a taxonomy. We used three prominent WordNet measures. With the procedure we applied on WordNet measures, they are individually not susceptible enough to the criterion of sentence similarity approximation. Therefore, based on the ratio, we dynamically choose several combinations of WordNet measures to capture the semantic similarity. To validate our proposed method, we conduct several experiments on the SemEval STS dataset. The experimental result revealed that our proposed method shows an efficient p erformance and outperformed some known related methods in terms of different evaluation metrics.