Tracking user interests over time is important for making accurate recommendations. However, the widely-used time-decay-based approach worsens the sparsity problem because it deemphasizes old item transactions. We introduce two ideas to solve the sparsity problem. First, we divide the users’ transactions into epochs i.e. time periods, and identify epochs that are dominated by interests similar to the current interests of the active user. Thus, it can eliminate dissimilar transactions while making use of similar transactions that exist in prior epochs. Second, we use a taxonomy of items to model user item transactions in each epoch. This well captures the interests of users in each epoch even if there are few transactions. It suits the situations in which the items transacted by users dynamically change over time; the semantics behind classes do not change so often while individual items often appear and disappear. Fortunately, many taxonomies are now available on the web because of the spread of the Linked Open Data vision. We can now use those to understand dynamic user interests semantically. We evaluate our method using a dataset, a music listening history, extracted from users’ tweets and one containing a restaurant visit history gathered from a gourmet guide site. The results show that our method predicts user interests much more accurately than the previous time-decay-based method.
A handover support system that supports care workers to share information and knowledge on patients and nursing-care work based on information recommendation is described. A handover is time consuming work because it takes much time to write and retrieve information on patients. We investigated the handover work in a nursing home, and found that about 25% of the work time was spent for sharing information among care workers. The aim of this study is to support care workers to share handover information efficiently.For this aim, we propose a novel handover support system called DANCE (Dynamic Action and kNowledge assistant for Collaborative sErvice fields) that supports care workers to share information and knowledge on patients and nursing-care work based on information recommendation. The system has following functions; (1) a function for recommending handover information based on attribute names and their values, (2) a function for recommending free-text contents of handover information, and (3) a function for sharing multimedia information. We had experiments for evaluating effectiveness of the system, and confirmed that the system can reduce the time for sharing handover information through a day compared to the time based on a notebook. We compared the work time for sharing two types of handover infomation between the system and notebook conditions; (a) information on patients and nursing-care work which is stored as pairs of attribute names and their values, (b) free-text contents on patients. Results of experiments revealed that the system can reduce the time for the former type of information as 55.2% (64.0s) per person a day compared to the notebook condition, and 59.0% (200s) for the latter type of information. An overview of the system and results of experiments are described.