Data envelopment analysis (DEA) has helped us in some many aspects to make our decision making a lot more accurate by providing better criteria selection when it comes to a selection of different combinations for different objectives. However, DEA is lacking of abilities that could observe a large amount of variables. To solve this problem and to identify the key points of successful businesses, this study introduces an advanced decision making scheme that combines stochastic neighbor embedding (SNE) and data envelopment analysis (DEA) to handle the performance measurement task. SNE will be used to condense large amounts into manageable data since SNE is a method that considers every point to be the neighbor with all the points. Any observation made with the data, the DEA will conduct the learning of every key point to reach a point of reference. In addition, this scheme will not only provide a precise evaluation outcome; it will also back it up with a forecast on capability by machine learning. The scheme promises and alternative performance forecasting and evaluation.
