MOVBOK: A Personalized Social Network Based Cross Domain Recommender System

Vinayak, Sharu; Sharma, Richa; Singh, Robert P.

doi:10.17485/ijst/2016/v9i31/96291

Cited by 6 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the term was coined in the early 90s, it became popular in 1997 with the important special issue of RS by Paul Resnik in Communication of the association for computing machinery (ACM), and as the result, there was a rise of RS continuously. RS was developed as an independent research field in the mid-1970s at Duke University [2]. They did this path of development not only but with the help of artificial intelligence (AI), information extraction, and human-computer interaction.…”

Section: Introductionmentioning

confidence: 99%

Provenance and social network analysis for recommender systems: a literature review

Rrmoku

Selimi²,

Ahmedi³

2022

IJECE

View full text Add to dashboard Cite

<span>Recommender systems (RS) and their scientific approach have become very important because they help scientists find suitable publications and approaches, customers find adequate items, tourists find their preferred points of interest, and many more recommendations on domains. This work will present a literature review of approaches and the influence that social network analysis (SNA) and data provenance has on RS. The aim is to analyze differences and similarities using several dimensions, public datasets for assessing their impacts and limitations, evaluations of methods and metrics along with their challenges by identifying the most efficient approaches, the most appropriate assessment data sets, and the most appropriate assessment methods and metrics. Hence, by correlating these three fields, the system will be able to improve the recommendation of certain items, by being able to choose the recommendations that are made from the most trusted nodes/resources within a social network. We have found that content-based filtering techniques, combined with term frequency-inverse document frequency (TF-IDF) features are the most feasible approaches when combined with provenance since our focus is to recommend the most trusted items, where trust, distrust, and ignorance are calculated as weight in terms of the relationship between nodes on a network.</span>

show abstract

Section: Introductionmentioning

confidence: 99%

Provenance and social network analysis for recommender systems: a literature review

Rrmoku

Selimi²,

Ahmedi³

2022

IJECE

View full text Add to dashboard Cite

show abstract

“…However, it is very tough to collect an abundant amount of labeled data for every real‐world scenario because of the insufficiency of easily available data or the high cost of data acquisition. Creating a labeled dataset for every possible domain is a costly and time‐consuming process 21 . Moreover, statistical classifiers presume that the training and testing data emanate from a shared essential distribution, but because of the large sparsity and variability of natural language, many times there are distribution discrepancies in the specialized training dataset and real‐world dataset 22 …”

Section: Introductionmentioning

confidence: 99%

“…With the aim to mitigate the dependency on a huge labeled dataset, domain adaptation has become a promising direction. Domain adaptation enables us to borrow the information from one domain to predict the performance in the other domain 21 . Cross‐domain sentiment classification (CDSC) is one of the subtask of sentiment analysis and utilizes techniques of domain‐adaptation to perform its tasks.…”

Section: Introductionmentioning

confidence: 99%

“…Creating a labeled dataset for every possible domain is a costly and time-consuming process. 21 Moreover, statistical classifiers presume that the training and testing data emanate from a shared essential distribution, but because of the large sparsity and variability of natural language, many times there are distribution discrepancies in the specialized training dataset and real-world dataset. 22 With the aim to mitigate the dependency on a huge labeled dataset, domain adaptation has become a promising direction.…”

mentioning

confidence: 99%

“…Domain adaptation enables us to borrow the information from one domain to predict the performance in the other domain. 21 Cross-domain sentiment classification (CDSC) is one of the subtask of sentiment analysis and utilizes techniques of domain-adaptation to perform its tasks. The task of CDSC is to train a classifier with a source domain labeled dataset, and then this classifier may be utilized for classification on the target domain.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Cross‐domain sentiment classification using decoding‐enhanced bidirectional encoder representations from transformers with disentangled attention

Singh

Sachan

Patel

2022

Concurrency and Computation

View full text Add to dashboard Cite

Cross-domain sentiment classification is a significant task of sentiment analysis that objectives to predict the opinion orientation of text documents in the target domain by using the source domain's learned classifier. Most of the existing approaches of domain-adaptation in sentiment classification focus on sharing low-dimensional features across the domain using domain independent and specific features to mitigate the gap between domains. Earlier cross-domain sentiment classification approaches mainly focused on document level and sentence level, they cannot consider the full impact of aspect words, position of the words, and long-term dependencies. To address this concern, we propose a model for cross-domain sentiment classification, which is based on decoding-enhanced BERT with disentangled attention (DeBERTa). DeBERTa is a pretrained language model based on transformer architecture. In this article, we perform sentence and aspect embedding to mine wordpiece information from text document. DeBERTa language-model utilize disentangled attention mechanism and an enhanced mask decoder to understand the expression features. Disentangled attention mechanism is used to encode each word into two vectors (i.e., content and position vector). In order to predict the masked tokens during model pretraining, an enhanced mask decoder is employed, which incorporates absolute positions in the decoding layer.Finally, experiments are conducted on the benchmark dataset that demonstrates the superiority of fine-tuned DeBERTa model for cross-domain sentiment classification tasks.

show abstract

Comparative Analysis of Music Genre Classification Framework Based on Deep Learning

Pattanaik,

Jain,

Sharma

et al. 2023

Algorithms for Intelligent Systems

View full text Add to dashboard Cite

MOVBOK: A Personalized Social Network Based Cross Domain Recommender System

Cited by 6 publications

References 22 publications

Provenance and social network analysis for recommender systems: a literature review

Provenance and social network analysis for recommender systems: a literature review

Cross‐domain sentiment classification using decoding‐enhanced bidirectional encoder representations from transformers with disentangled attention

Comparative Analysis of Music Genre Classification Framework Based on Deep Learning

Contact Info

Product

Resources

About