Abstract:Many businesses benefit from the business-to-business (B2B) campaign recommendation strategy. Complex commodities and user profiles restrict the performance of automated B2B marketing recommendations to business owners. The existing solutions supplied B2B marketing campaign solutions without accounting for social and ambiguous preferences, limiting the accuracy of B2B marketing campaign recommendations. It motivates us to propose a novel model to improve the performances further. To achieve accurate and fast B2B marketing campaign recommendations, we design a novel approach consisting of social context-aware (SCA), fuzzy preference (FP), personalized temporal graph (PTG), and low-rank graph reconstruction (LRGR). The proposed SCA-FTL approach consolidates the social and fuzzy preferences for accurate recommendations. The main steps of the SCA-FTL model are pre-processing, preference modeling, and automatic recommendations. Pre-processing removes noise and artifacts to improve suggestion accuracy. The hybrid preference extraction uses FP-PTG and LRGR. PF-PTG describes each client’s purchasing strategy using a fuzzy methodology to solve ambiguous/vague tailored preferences. Mixing FP-PTG with social context helps derive customer preferences. SCA-FT gets each customer’s preferences using the fuzzy temporal graph and social context. LRGR discovers unobserved edges in SCA-FT to increase recommendation predictability. SCA-FT-LRGR findings provide B2B marketing ideas. The results of the experiments reveal that the proposed model recommended B2B marketing campaigns with improved accuracy and minimum computing requirements than the existing solutions. Compared to existing methods, overall recommendation accuracy improves by 25.23%, and recommendation time decreases by 8.46%.

Social context-aware and fuzzy preference temporal graph for personalized B2B marketing campaigns recommendations

Improved Metric-Based Recommender by Historical Interactions.

Exploiting Structural and Temporal Influence for Dynamic Social-Aware Recommendation

Graph Based Recommendations: From Data Representation to Feature Extraction and Application

GRAINRec: Graph and Attention Integrated Approach for Real-Time Session-Based Item Recommendations

Incorporating User Preferences Across Multiple Topics Into Collaborative Filtering For Personalized Merchant Recommendation

A Personalized Collaborative Filtering Recommendation System Based on Bi-Graph Embedding and Causal Reasoning

Learning Personalized Page Content Ranking Using Customer Representation

Diversifying Personalized Recommendation with User-session Context

Personalized Recommendation by Exploring Social Users' Behaviors

Personalized recommendation: From clothing to academic

Personalized Session-Based Recommendation Using Graph Attention Networks.

Enhancing Session-Based Social Recommendation Through Item Graph Embedding and Contextual Friendship Modeling.

A Multiple Criteria Approach Integrating Social Ties to Support Purchase Decision

Personalized Promotion Recommendation Through Consumer Experience Evolution Modeling

Integration of short and long-term interests: A preference aware session-based recommender

FSBPR: a novel approach to improving BPR for recommendation with the fusion of similarity

Time-aware Hyperbolic Graph Attention Network for Session-based Recommendation

Towards Personalized Federated Multi-Scenario Multi-Task Recommendation

A general graph-based framework for top-N recommendation using content, temporal and trust information

An Interactive Personalized Recommendation System Using the Hybrid Algorithm Model