Ageing is a very complex process leading to the degradation of health conditions in living beings. We aim to classify the proteins which are related to ageing and non-ageing. This might help us further to understand the mechanism of healthy ageing in living beings which will be a great contribution to this decade which has been titled as ‘Decade of Healthy ageing’. The model has been built using a dataset containing of about 20183 proteins along with 21001 protein features. A deep learning-based technique has been used for classification of protein as ageing or non-ageing based on proteins in GenAge dataset. We label the Human proteins of Swiss-prot as ‘0’ or ‘1’. ‘0’ indicates it is a non-ageing protein and ‘1’ indicates as ageing proteins as per the GenAge database. Using neural network, we have developed a model for the prediction and classification of ageing and non-ageing proteins. Our developed model predicts the probabilistic value for all proteins in the dataset. Higher the prediction probability value, higher is the probability that it is an ageing protein. An accuracy of 98.41% was obtained and identification of the new proteins as ageing/non-ageing has been made possible which also have strong computational evidence in ageing Based on previous studies, the proteins which have been already identified as ageing proteins has also been classified as ageing protein by our developed model with the higher probability value for example, as in the case of ‘Cellular tumor antigen p53 (P53_HUMAN)’. Our model also helped to obtain some new proteins implicated in ageing with strong evidence of their important role in ageing such as ‘Proto-oncogene tyrosine-protein kinase Src (SRC_HUMAN)’, ‘Heat shock protein HSP 90-beta (HS90B_HUMAN)’.