With the development of online communities, exploring and publishing of datasets and collaborative development with data scientists and machine learning engineers is exponentially increasing. The participation in competitions to solve data science challenges and the formation of machine learning communities has matured. Kaggle is one of the well-known machine learning hub to explore competitions, datasets, notebooks, and forums. Knowing the evolution and the mere future of datasets, competitions, and courses would benefit not only for users of Kaggle but also Kaggle administrators. This paper aims to develop a novel design pattern-based decision support system using a sequence model, Auto Regressive Integrated Moving Average (ARIMA) model to predict the total number of competitions, the number of competitions in the top 5 tags and the number of datasets using sequence models. Experiments are conducted using the Meta Kaggle dataset to evaluate the performance of the design pattern-based ARIMA model and the results show that the proposed forecasting model provides significant results in terms of MAPE (Mean Absolute Percentage Error) and other design pattern-based evaluation measures. Experimental results of the proposed system are compared with Long-Short Term Memory (LSTM) based models and the results reveal that the proposed system is superior to LSTM based system in terms of minimizing the error.