Academic Articles Recommendation Using Concept-Based Representation

Mohamed, Dina; El-Kilany, Ayman; Mokhtar, Hoda M. O.

doi:10.1007/978-3-030-55187-2_52

Cited by 1 publication

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results are shown in Fig. 9 illustrate that without building a researcher's profile, the average recall of the proposed hierarchical document representation model is better than both the concept-based model [10] and the LDA+Word2vec model [9]. The proposed model improves the results of the recommendation systems for the dataset of size 6000 papers with 9% from the results of the concept-based model, and with 25% from the results of the (LDA+Word2vec) model.…”

Section: Discussionmentioning

confidence: 91%

“…In [28], the authors applied a multi labels classification for news articles where the word2vec model was applied to build a vector for words in news articles to capture the similarity between the words and then use those words vectors as a classification feature. The authors in [10] proposed a model for representing academic articles to recommend them to researchers. This method generates a set of concepts by clustering the word vectors that are learned from the word2vec model where the words with the same semantic meaning will be grouped in one concept.…”

Section: Related Workmentioning

confidence: 99%

“…We compared our model against the concept-based model [10] and the LDA+Word2vec model [9] given their similarity with the proposed model. Both of concept-based model and the LDA+Word2vec model applied the word2vec model to represent the words.…”

Section: Performance Evaluationmentioning

confidence: 99%

“…Also, the number of concepts in the concept-based model was set to 50. Whereas the vector size 50 is chosen based on the experiments that have been conducted for the concept-based model [10] for different sizes of the dataset with different vector sizes (10,30, and 50), while the best results were achieved using the vector of size 50.…”

Section: Performance Evaluationmentioning

confidence: 99%

“…Then, a subset of each researcher's preferred papers was aggregated as the researcher profile using their representations. Researchers' profiles were used to recommend papers to the researcher where the recommendation results outperform other semanticbased representation models like LDA with word2vec combination [9], and concept-based representation [10].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Hybrid Model for Documents Representation

Mohamed¹,

El-Kilany²,

Mokhtar³

2021

IJACSA

Self Cite

View full text Add to dashboard Cite

Text representation is a critical issue for exploring the insights behind the text. Many models have been developed to represent the text in defined forms such as numeric vectors where it would be easy to calculate the similarity between the documents using the well-known distance measures. In this paper, we aim to build a model to represent text semantically either in one document or multiple documents using a combination of hierarchical Latent Dirichlet Allocation (hLDA), Word2vec, and Isolation Forest models. The proposed model aims to learn a vector for each document using the relationship between its words' vectors and the hierarchy of topics generated using the hierarchical Latent Dirichlet Allocation model. Then, the isolation forest model is used to represent multiple documents in one representation as one profile to facilitate finding similar documents to the profile. The proposed text representation model outperforms the traditional text representation models when applied to represent scientific papers before performing contentbased scientific papers recommendation for researchers.

show abstract

Section: Discussionmentioning

confidence: 91%

Section: Related Workmentioning

confidence: 99%