Using Context to Improve Predictive Modeling of Customers in Personalization Applications

Abstract

Abstract-The idea that context is important when predicting customer behavior has been maintained by scholars in marketing and data mining. However, no systematic study measuring how much the contextual information really matters in building customer models in personalization applications has been done before. In this paper, we study how important the contextual information is when predicting customer behavior and how to use it when building customer models. It is done by conducting an empirical study across a wide range of experimental conditions. The experimental results show that context does matter when modeling the behavior of individual customers and that it is possible to infer the context from the existing data with reasonable accuracy in certain cases. It is also shown that significant performance improvements can be achieved if the context is "cleverly" modeled, as described in this paper. These findings have significant implications for data miners and marketers. They show that contextual information does matter in personalization and companies have different opportunities to both make context valuable for improving predictive performance of customers' behavior and decreasing the costs of gathering contextual information.

Trust beyond reputation: A computational trust model based on stereotypes

“…However, from the perspective of behavior science [11], as well as being supported by recent works that use contextual information to predict user behavior in various information systems [30,27,29], we believe that an agent's behavior change in the transactions is correlated with and can be inferred (to certain extent) by the associated contextual information (e.g., by considering the dynamic trust [21,23]). For instance, in an online auction site like eBay, a seller may vary his behavior consciously or unwittingly in selling different items (e.g., he may be careful when selling expensive goods, but imprudent with cheap ones).…”

Mobile Movie Recommendations with Linked Data

“…As in [20], we assume that there is a predefined finite set of contextual types in a given application and each of these types has a well-defined structure. In particular, in our mobile scenario we consider the context as represented by the following information:…”

A Framework for Time-Aware Recommendations

“…Time is modeled here as a multidimensional attribute. The dimensions of time have a hierarchical structure, that is, time values are organized at different levels of granularity (similar to [16,18]). In particular, we consider three different levels over time: time of day, day of week and time of week with domain values {"morning", "afternoon", "evening", "night"}, {"Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"} and {"Weekday", "Weekend"}, respectively.…”

“…Several extensions have been proposed, such as employing multi-criteria ratings (e.g., [2]) and defining recommendations for groups (e.g., [4,15,14]). Recently, there are also approaches focusing on enhancing recommendations with further contextual information (e.g., [3,16]). In these approaches, context is defined as a set of dimensions, or attributes, such as location, companion and time, with hierarchical structure.…”