Long Short-Term Memory (LSTM) to Predict the Viewability of any Page Depth for any Given Dwell Time

Abstract

In online distributers had a one major income source that is displaying advertising through online. In existing techniques recommender systems are depending upon the user’s interests. Recent studies show that the ads were really not seen by user’s means they don't scroll sufficiently profound to get the advertisements see. For this reason a new model was discovered for advertisements are paid on the off chance that they are in view, not minimally being served. A critical issue for distributers be near expect the chance to an advertisement on a agreed sheet intensity motivation live appeared resting on a client's monitor intended for a convinced live instance. This manuscript suggests Long Short-Term Memory (LSTM) near forecast the perceptibility of every sheet intensity intended for every agreed abide moment. It is a arrangement of bi-directional LSTM networks, encoder decoder structures & outstanding associations. The consequences shows that the high performance in terms of prediction.