Building intelligent shopping assistants using individual consumer models

Abstract

ThispaperdescribesanIntelligentShoppingAssistantdesigned for a shopping cart mounted tablet PC that enables individual interactions with customers. We use machine learning algorithms to predict a shopping list for the customer's current trip and present this list on the device. As they navigate through the store, personalized promotions are presented using consumer models derived from loyalty card data for each inidvidual. In order for shopping assistant devices to be effective, we believe that they have to be powered by algorithms that are tuned for individual customers and can make accurate predictions about an individual's actions. We formally frame the shopping list prediction as a classification problem, describe the algorithms and methodology behind our system, and show that shopping list prediction can be done with high levels of accuracy, precision, and recall. Beyond the prediction of shopping lists we briefly introduce other aspects of the shopping assistant project, such as the use of consumer models to select appropriate promotional tactics, and the development of promotion planning simulation tools to enable retailers to plan personalized promotions delivered through such a shopping assistant.

Self-configuring data mining for ubiquitous computing

“…The following list ranks the decision tree models by their score: (8,12,2,6,15,5,14,10,9,13,1,4,3,7,11) Final observation supporting pre-screening presumption is that, five out of six class label attributes that are most likely to be eliminated by prescreening (first six bars in Figure 8) are among the class label attributes of six worst scored decision tree models. Hence, it is possible to say that prescreening eliminates the decision trees that are very unlikely to be selected as the appropriate model for configuration decisions by Algorithm 2.…”

“…Let Q(qm : A,9ii2 : A,<?i2i • A,<?i22 : A,<?i3 : D 5 ,q 2U '• D 6 ,q 2i2 • D 7 ,q 22 : A) be the relation schema defining the quality attributes in the execution file of the experiment.…”

Web-Enabled and Improved Software Tools and Data Are Needed to Measure Nutrient Intakes and Physical Activity for Personalized Health Research

“…Whereas a single food store visit may not be an accurate assessment of nutrient intakes for a household, food purchases for a longer period of time (.3-4 mo) may provide valuable data for assessing nutrient intakes. To mitigate the obvious limitation that purchases are for a household rather than an individual, algorithms are available that associate food purchases with age, sex, income status, and other demographic factors across tens of thousands of individuals (51). For example, one store may have over 2000 customers and 150,000 transactions within a year (3).…”

Predictive text input in a mobile shopping assistant

“…There are two kinds of mobile shopping assistants: those for shopping malls [2,4] and those for individual shops [3,12]. Shopping assistants have been envisaged for navigation support (both finding shops in malls and products within shops), promotions, and many other functions [3,7].…”

“…Shopping assistants have been envisaged for navigation support (both finding shops in malls and products within shops), promotions, and many other functions [3,7]. Assistants for individual shops tend to focus on product level information whereas mall assistants tend to focus on navigation and shop locating features.…”