This report presents a comparative analysis of ChatGPT and DeepSeek AI based on synthetic data sourced from Kaggle. The study evaluates various user experience metrics, device compatibility, and time series analysis to determine which AI platform is more effective. Key findings indicate that DeepSeek AI outperforms ChatGPT in user rating, engagement, and response accuracy, while ChatGPT provides faster response times. The report further discusses statistical significance tests and forecasting using the Prophet library, highlighting an increasing user trend for DeepSeek AI over time. These insights can help users and organizations make informed decisions regarding AI chatbot selection.
This report compares two AI chatbots, ChatGPT and DeepSeek AI, focusing on their effectiveness in user engagement and experience. The analysis incorporates user ratings, session durations, response accuracy, and speed. Additionally, device compatibility and time series analysis are examined to assess long-term trends. The study aims to identify strengths and weaknesses of both platforms, providing valuable insights for AI users and developers.
Numerous studies have evaluated AI chatbots in terms of user engagement, performance, and adoption trends. Research on conversational AI suggests that response accuracy, speed, and overall user experience significantly impact user retention. Prior comparisons between AI models have also emphasized trade-offs between efficiency and accuracy, aligning with findings in this report. Furthermore, time series analysis has been widely used in forecasting AI adoption trends, making it a relevant methodology for this study.
The analysis is based on synthetic data obtained from Kaggle, ensuring a structured evaluation of ChatGPT and DeepSeek AI. The following methods were employed:
Descriptive Statistics: Comparison of mean values for key metrics such as user rating, experience score, session duration, response accuracy, and speed.
Device Compatibility Analysis: Evaluation of chatbot usability across different devices (laptop, mobile, smart speakers, and tablets).
Time Series Analysis: Forecasting trends in user engagement using the Prophet library.
Statistical Testing: Seasonal decomposition applied to validate trends.
User Experience Analysis: The study measured key performance indicators (KPIs) for both AI platforms.
Device Compatibility Testing: Data was segmented based on device types to determine the preferred platform for each.
Trend Forecasting: A time series model was applied to predict future chatbot engagement.
Significance Testing: Statistical tests ensured that observed differences were not due to randomness.
User Experience Metrics: DeepSeek received higher ratings (4.80 vs. 3.99) and longer session durations (34.69s vs. 22.56s) compared to ChatGPT. Response accuracy was also higher (0.89 vs. 0.80), while ChatGPT responded faster (3.44s vs. 1.24s).
Device Compatibility: ChatGPT was preferred on laptops and mobiles, whereas DeepSeek excelled on smart speakers and tablets.
Time Series Analysis: DeepSeek's user base is growing, while ChatGPT is experiencing a decline.
The analysis reveals DeepSeek’s superiority in engagement and accuracy, making it more suitable for users prioritizing reliability. ChatGPT, on the other hand, offers faster responses, appealing to users valuing quick interactions. Device compatibility findings suggest that platform choice may depend on user preferences for specific device categories. The forecasting results indicate a promising future for DeepSeek, necessitating further real-world validation.
DeepSeek AI demonstrates superior performance in user experience metrics, accuracy, and engagement, making it a preferred choice for sustained interactions. ChatGPT, however, maintains an edge in response speed. Organizations must weigh these factors based on their specific requirements. Future research should incorporate real-world datasets to validate these findings.
Kaggle dataset: DeepSeek vs ChatGPT AI Platform Comparison
Research papers on conversational AI and chatbot user engagement
Documentation for Prophet time series analysis
We extend our gratitude to Kaggle for providing the dataset, as well as to the developers of ChatGPT and DeepSeek AI for enabling this comparative study.
Table of Mean Values: Includes numerical data for user ratings, session durations, response accuracy, etc.
Charts & Graphs: Visual representations of findings, including user engagement trends and statistical comparisons.
Dataset Description: Detailed information about the dataset structure and key variables used in the study.