From ChatGPT-3 to GPT-4: A Significant Leap in AI-Driven NLP Tools

Ahsan, Mohebul; Rahaman, Md. Saidur; Anjum, Nishath

doi:10.2139/ssrn.4404397

Cited by 4 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reviewed studies have taken diverse approaches to using LLMs for mental health classification in social media posts. Earlier works relied on transformer models like BERT (see, e.g., Devika et al, 2022;Yang et al, 2023), while more recent studies leverage advanced LLMs like GPT-3 and ChatGPT (Ahsan et al, 2023). The techniques also differ, with combinations such as BERT-CNN (Yang et al, 2023) and metapath2vec-doc2vec explored (Dong et al, 2017).…”

Section: Comparison Of Research Direction With Previous Workmentioning

confidence: 99%

See 1 more Smart Citation

Predictive Analytics in Mental Health Leveraging LLM Embeddings and Machine Learning Models for Social Media Analysis

Radwan,

Amarneh,

Alawneh

et al. 2024

International Journal of Web Services Research

View full text Add to dashboard Cite

The prevalence of stress-related disorders has increased significantly in recent years, necessitating scalable methods to identify affected individuals. This paper proposes a novel approach utilizing large language models (LLMs), with a focus on OpenAI's generative pre-trained transformer (GPT-3) embeddings and machine learning (ML) algorithms to classify social media posts as indicative or not of stress disorders. The aim is to create a preliminary screening tool leveraging online textual data. GPT-3 embeddings transformed posts into vector representations capturing semantic meaning and linguistic nuances. Various models, including support vector machines, random forests, XGBoost, KNN, and neural networks, were trained on a dataset of >10,000 labeled social media posts. The top model, a support vector machine, achieved 83% accuracy in classifying posts displaying signs of stress.

show abstract

Section: Comparison Of Research Direction With Previous Workmentioning

confidence: 99%

“…It is widely used for classification problems. KNN (Ahsan et al, 2023) predicts new data points based on similarity to the k-nearest neighbors in the training data. It assumes proximity implies similar outputs.…”

Section: Comparative Experimentsmentioning

confidence: 99%

Predictive Analytics in Mental Health Leveraging LLM Embeddings and Machine Learning Models for Social Media Analysis

Radwan,

Amarneh,

Alawneh

et al. 2024

International Journal of Web Services Research

View full text Add to dashboard Cite

show abstract

“…Even though GPT-4 does not officially have any personality traits, GPT-4 can be made to generate discriminatory texts: "GPT-4 is capable of generating discriminatory content favorable to autocratic governments across multiple languages" [2] (p. 51). Ahsan et al [22] examined GPT-4 for truthfulness and came to the conclusion that GPT-4 gives significantly better answers in terms of quality compared with GPT-3 [22] (p. 3). However, according to OpenAI [2] (p. 10), it is important to know that GPT-4 is not always completely reliable.…”

Section: Personality Of Chatgptmentioning

confidence: 99%

The Personification of ChatGPT (GPT-4)—Understanding Its Personality and Adaptability

Stöckli,

Joho,

Lehner

et al. 2024

Information

View full text Add to dashboard Cite

Thanks to the publication of ChatGPT, Artificial Intelligence is now basically accessible and usable to all internet users. The technology behind it can be used in many chatbots, whereby the chatbots should be trained for the respective area of application. Depending on the application, the chatbot should react differently and thus, for example, also take on and embody personality traits to be able to help and answer people better and more personally. This raises the question of whether ChatGPT-4 is able to embody personality traits. Our study investigated whether ChatGPT-4’s personality can be analyzed using personality tests for humans. To test possible approaches to measuring the personality traits of ChatGPT-4, experiments were conducted with two of the most well-known personality tests: the Big Five and Myers–Briggs. The experiments also examine whether and how personality can be changed by user input and what influence this has on the results of the personality tests.

show abstract

“…This version boasts an impressive parameter count of 170 trillion, positioning it as one of the most sophisticated chatbots developed thus far (Hanna & Levic, 2023;. Ahsan et al (2023) give exhaustive clarifications of the improvements made to the latest cycle of ChatGPT. These improvements include the reconciliation of state-of-the-art profound learning strategies, an upgraded fake brain network structure, and an expanded multimodal language model fit for breaking down text-based and visual data.…”

Section: Introductionmentioning

confidence: 99%

ChatGPT's Applications in Higher Education: Unmasking Opportunities and Challenges

Sarker,

Eiaz-Ur-Rahman,

Sakib

et al. 2023

J. Educ., Mngt., and Dev. Studies

View full text Add to dashboard Cite

The emergence of generative artificial intelligence (AI), as demonstrated by models like ChatGPT, has garnered significant attention in recent times owing to its capacity to transform multiple domains of society, including higher education. Within academia, individuals such as students, educators, and scholars have already employed Large Language Models (LLMs) such as ChatGPT for a diverse range of academic and non-academic endeavors. However, there is still debate regarding the use of ChatGPT in higher education. The acceptability and validity of its impact on higher education have been a recent burning issue. The primary aim of this review paper is to elucidate the utilization of ChatGPT within the context of higher education, considering its possible opportunities and challenges. The data utilized in this study has been generated by examining published papers, websites, and blogs. Using an "introductory literature review," we have analyzed, synthesized, and explained the data that was generated. The findings revealed inadequate technological access, overreliance on AI, the presence of inaccuracies within subject-based knowledge, and the inability of educators to adapt teaching strategies are the core challenges of ChatGPT in higher education. However, besides these challenges, ChatGPT also brings opportunities like the advancement of learning efficiency, the concept of personalized learning, support for educators, and opportunities for skill enhancement.

show abstract

From ChatGPT-3 to GPT-4: A Significant Leap in AI-Driven NLP Tools

Cited by 4 publications

References 0 publications

Predictive Analytics in Mental Health Leveraging LLM Embeddings and Machine Learning Models for Social Media Analysis

Predictive Analytics in Mental Health Leveraging LLM Embeddings and Machine Learning Models for Social Media Analysis

The Personification of ChatGPT (GPT-4)—Understanding Its Personality and Adaptability

ChatGPT's Applications in Higher Education: Unmasking Opportunities and Challenges

Contact Info

Product

Resources

About