Detecting fake news and disinformation using artificial intelligence and machine learning to avoid supply chain disruptions

Abstract

Fake news and disinformation (FNaD) are increasingly being circulated through various online and social networking platforms, causing widespread disruptions and influencing decision-making perceptions. Despite the growing importance of detecting fake news in politics, relatively limited research efforts have been made to develop artificial intelligence (AI) and machine learning (ML) oriented FNaD detection models suited to minimize supply chain disruptions (SCDs). Using a combination of AI and ML, and case studies based on data collected from Indonesia, Malaysia, and Pakistan, we developed a FNaD detection model aimed at preventing SCDs. This model based on multiple data sources has shown evidence of its effectiveness in managerial decision-making. Our study further contributes to the supply chain and AI-ML literature, provides practical insights, and points to future research directions.

Fake News Detection by Fine Tuning of Bidirectional Encoder Representations from Transformers

“…Bi LSTM [21] 90.40% Random Forest [19] 84.40% Recurrent neural networks and transformer based models [20] 95% (f1-score) TFIDF with CNN [18] 96.89% FND NS [14] 74.8% SVM [8] 94.19% LSTM with Glove embeddings [17] 98.6% Proposed method 99.09% data, which includes everything from eliminating raw data to lemmatization for all three datasets. Machine learning models rely on the TF-IDF for feature analysis, whereas deep learning models depend on word embedding.…”

“…They reported that the SVM approach achieved the best results with 94.19% accuracy. Similarly, SVM classifier has been examined for predicting fake news using dimensionality reduction in [8].…”

Integration of machine learning in the supply chain for decision making: A systematic literature review

“…• Natural language processing techniques (NLP) collect and synthesize information using the Support Vector Machine (SVM) (Akhtar et al, 2022) Cost and financial management…”

Fake news research trends, linkages to generative artificial intelligence and sustainable development goals

“…ChatGPT operates as a text-based conversational agent, providing textual responses to user queries [ 59 ]. AI algorithms have been shown to be useful in detecting fake news or misinformation that may be interfering with efficiency and optimization [ 60 , 61 ]. Proponents of using AI in the detection of fake news suggest that certain principles need to be followed, including the development of strategies by the software designers to combat fake news, enabling software users to report fake news when detected, and keeping users informed of the dissemination of fake news [ 62 ].…”