Modified node2vec and attention based fusion framework for next POI recommendation

Kumar, Akshi; Jain, Deepak Kumar; Mallik, Abhishek and Kumar, Sanjay. 2024. Modified node2vec and attention based fusion framework for next POI recommendation. Information Fusion, 101, 101998. ISSN 1566-2535 [Article]

No full text available
[img] Text
elsarticle-template-1-infusion_R1_without_mark_changes.pdf - Accepted Version
Permissions: Administrator Access Only until 1 March 2025.
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (755kB)

Abstract or Description

The rise of location-based services has led to the widespread adoption of location-based social networks (LBSNs), which play a vital role in making recommendations for the next Point-of-Interest (POI). This paper introduces a modified node2Vec and attention-based fusion framework for the next POI recommendation. We start by preprocessing the raw data to gather the relevant information and present a modified node2vec algorithm to generate the feature vectors for users and locations. These feature vectors are then processed using the attention-based framework. The processed features are then used to create well-labeled and balanced datasets which are grouped by specific time intervals. These datasets are then used for training various ML classifiers which are ensembled in a weighted manner to make an improved fusion based recommendation system. The intensive experimental simulations demonstrate the effectiveness of the proposed framework over existing state-of-art methods.

Item Type:

Article

Identification Number (DOI):

https://doi.org/10.1016/j.inffus.2023.101998

Data Access Statement:

Data will be made available on request.

Keywords:

Attention-based framework, Ensemble learning, Information Fusion, Next Point-Of-Interest recommendation system, Location-based Social Networks (LSBNs), Modified node2vec embedding

Departments, Centres and Research Units:

Computing

Dates:

DateEvent
28 August 2023Accepted
8 September 2023Published Online
January 2024Published

Item ID:

34074

Date Deposited:

16 Oct 2023 12:26

Last Modified:

17 Oct 2023 07:31

Peer Reviewed:

Yes, this version has been peer-reviewed.

URI:

https://research.gold.ac.uk/id/eprint/34074

View statistics for this item...

Edit Record Edit Record (login required)