The Kepler space telescopic has created a huge quantity of astronomical time series data that has completely changed the comprehension of explanatory systems. Light curves constitute scanty, sparse, and unpredictable representations of a star’s illumination over time. It is necessary to complete a process of classification using these enormous time series information in order to identify false positives as well as planet candidates. Neural networks, a type of deep learning technology, have shown promise in effectively distinguishing potential astrophysical illuminating candidates among different occurrences like star fluctuations and consistent instrumental influence. The Kepler Space Mission’s main goal intended to investigate the broad range and organization of the solar system. This paper presents a methodology using a Convolutional Neural Network (CNN) for the automatic identification of exoplanets in the time series information obtained through the Kepler Space Telescope. The method divides transit information into exoplanets and non-exoplanets utilizing various methods of signal prior processing alongside a properly trained deep learning model. The framework has outstanding performance measurements, such as recall, accuracy, as well as precision. The precision-recall curve, which offers thorough insights into the model’s prediction skills across different probability restrictions, is a crucial component of the research. These results highlight the durability and efficiency of the model when performing the exoplanet classification task.