We propose an algorithm for learning hierarchical user interest models according to the Web pages users have browsed. In this algorithm, the interests of a user are represented into a tree which is called a user inter...We propose an algorithm for learning hierarchical user interest models according to the Web pages users have browsed. In this algorithm, the interests of a user are represented into a tree which is called a user interest tree, the content and the structure of which can change simultaneously to adapt to the changes in a user＇s interests. This expression represents a user＇s specific and general interests as a continuurn. In some sense, specific interests correspond to shortterm interests, while general interests correspond to longterm interests. So this representation more really reflects the users＇ interests. The algorithm can automatically model a us er＇s multiple interest domains, dynamically generate the in terest models and prune a user interest tree when the number of the nodes in it exceeds given value. Finally, we show the experiment results in a Chinese Web Site.展开更多
Modelling of the agglomeration and deposition on a constricted tube collector of colloidal size particles immersed in a liquid is investigated using the discrete element method （DEM）. The ability of this method to r...Modelling of the agglomeration and deposition on a constricted tube collector of colloidal size particles immersed in a liquid is investigated using the discrete element method （DEM）. The ability of this method to represent surface interactions allows the simulation of agglomeration and deposition at the particle scale. The numerical model adopts a mechanistic approach to represent the forces involved in colloidal suspensions by including near-wall drag retardation, surface interaction and Brownian forces. The model is implemented using the commercially available DEM package EDEM 2.3~, so that results can be repli- cated in a standard and user-friendly framework. The effects of various particle-to-collector size ratios, inlet fluid flow-rates and particle concentrations are examined and it is found that deposition efficiency is strongly dependent on the inter-relation of these parameters. Particle deposition and re-suspension mechanisms have been identified and analyzed thanks to EDEM＇s post processing capability. One-way coupling with computational fluid dynamics （CFD） is considered and results are compared with a two- way coupling between EDEM 2.3 and FLUENT 12.1. It is found that two-way coupling requires circa 500% more time than one-way coupling for similar results.展开更多
基金Supported by the National Natural Science Funda-tion of China (69973012 ,60273080)
文摘We propose an algorithm for learning hierarchical user interest models according to the Web pages users have browsed. In this algorithm, the interests of a user are represented into a tree which is called a user interest tree, the content and the structure of which can change simultaneously to adapt to the changes in a user＇s interests. This expression represents a user＇s specific and general interests as a continuurn. In some sense, specific interests correspond to shortterm interests, while general interests correspond to longterm interests. So this representation more really reflects the users＇ interests. The algorithm can automatically model a us er＇s multiple interest domains, dynamically generate the in terest models and prune a user interest tree when the number of the nodes in it exceeds given value. Finally, we show the experiment results in a Chinese Web Site.
文摘Modelling of the agglomeration and deposition on a constricted tube collector of colloidal size particles immersed in a liquid is investigated using the discrete element method （DEM）. The ability of this method to represent surface interactions allows the simulation of agglomeration and deposition at the particle scale. The numerical model adopts a mechanistic approach to represent the forces involved in colloidal suspensions by including near-wall drag retardation, surface interaction and Brownian forces. The model is implemented using the commercially available DEM package EDEM 2.3~, so that results can be repli- cated in a standard and user-friendly framework. The effects of various particle-to-collector size ratios, inlet fluid flow-rates and particle concentrations are examined and it is found that deposition efficiency is strongly dependent on the inter-relation of these parameters. Particle deposition and re-suspension mechanisms have been identified and analyzed thanks to EDEM＇s post processing capability. One-way coupling with computational fluid dynamics （CFD） is considered and results are compared with a two- way coupling between EDEM 2.3 and FLUENT 12.1. It is found that two-way coupling requires circa 500% more time than one-way coupling for similar results.
