Consequently, if a targeted client has a plan to only visit one or two stores, then, AdNext will not be able to make any predictions.
Instead of predicting the business type, our prediction algorithm, P-DPA, predicts a set of stores that could be visited by a targeted client, and then, uses the predicted stores to generate a list of advertisements for the targeted client.
In addition, P-DPA algorithm can predict the first store to visit by a targeted client while AdNext will need to wait for the targeted client to visit at least two stores in order to be able to make any predictions.
AdNext uses the business type of the last two locations visited by a targeted client to predict the business type of the next location for that client.
Kim et al. [3] developed a system called, AdNext, that uses Bayesian networks to build a transition matrix for the shopping mall clients in order to predict the business type of the next location that a targeted client could visit.
Another group of researchers designed a framework called LASA, Location Aware Shopping Advertisement, that uses an ontology based formulation of clients and products profiles to generate a list of ads related to the selection history of a targeted client [5].
Consequently, failing to discover the correct location of a targeted client will hinder the process of determining the list of products' titles that should be sent to that client.
In addition, our P-DPA algorithm predicts a set of stores that could be visited and then generate a list of ads for those stores instead of just generating a possibly long list of ads for every store in the discovered area of a targeted client as in LASA.
In addition, AdNext detects the current location of the targeted client by using the shopping mall's access points.
Then, the client can click on any of the predicted stores and a list of advertisements will be generated and sent to him/her, as shown in Fig. 25. Figure 26 shows the details of a chosen ad by the targeted client.
Furthermore, little is known about the targeted client's understanding of mobile money, what is their convenience level, do they consider as safe transaction, the challenges they faced, and what they think about the ways to improve it.
