【摘 要】
：
Relevance estimation is one of the core concerns of information retrieval (IR) studies.Although existing retrieval models gained much success in both deepening our under-standing of information seeking behavior and building effec-tive retrieval systems,we
【机 构】
：
Department of Computer Science and Technology,Institute for Artificial Intelligence,Beijing National
论文部分内容阅读
Relevance estimation is one of the core concerns of information retrieval (IR) studies.Although existing retrieval models gained much success in both deepening our under-standing of information seeking behavior and building effec-tive retrieval systems,we have to admit that the models work in a rather different manner from how humans make relevance judgments.Users\' information seeking behaviors involve com-plex cognitive processes,however,the majority of these behav-ior patterns are not considered in existing retrieval models.To bridge the gap between practical user behavior and retrieval model,it is essential to systematically investigate user cogni-tive behavior during relevance judgement and incorporate these heuristics into retrieval models.In this paper,we aim to for-mally define a set of basic user reading heuristics during rel-evance judgement and investigate their corresponding model-ing strategies in retrieval models.Further experiments are con-ducted to evaluate the effectiveness of different reading heuris-tics for improving ranking performance.Based on a large-scale Web search dataset,we find that most reading heuristics can im-prove the performance of retrieval model and establish guide-lines for improving the design of retrieval models with human-inspired heuristics.Our study sheds light on building retrieval model from the perspective of cognitive behavior.
其他文献
Advanced model-based control strategies,e.g.,model predictive control,can offer superior control of key process variables for multiple-input multiple-output sys-tems.The quality of the system model is critical to controller performance and should adequate
To study the dynamic behavior of a process,time-resolved data are collected at different time instants during each of a series of experiments,which are usually designed with the design of experiments or the design of dynamic experiments methodologies.For
The catalytic hydrogenation of carboxylic acid to alcohols is one of the important strategies for the conversion of biomass.Herein,a series of Ni-doped PtSn catalysts were prepared,characterized and studied in the hydrogenation of acetic acid.The Ni dopan
Solvent-based post-combustion capture tech-nologies have great potential for CO2 mitigation in traditional coal-fired power plants.Modelling and simula-tion provide a low-cost opportunity to evaluate perfor-mances and guide flexible operation.Composed by
Simulation is besides experimentation the major method for designing,analyzing and optimizing chemical processes.The ability of simulations to reflect real process behavior strongly depends on model quality.Validation and adaption of process models are us
Compared to conventional hyperthermia that is limited by low selectivity and severe side effects,nano-enabled hyperthermia yields great potentials to tackle these limitations for cancer treatment.Another major advance is the observation of immunological r
Dimethyl carbonate is an eco-friendly essential chemical that can be sustainably produced from CO2,which is available from carbon capture activities or can even be captured from the air.The rapid increase in dimethyl carbonate demand is driven by the fast
Flowsheet simulations of chemical processes on an industrial scale require the solution of large systems of nonlinear equations,so that solvability becomes a practical issue.Additional constraints from technical,economic,environmental,and safety considera
There are many optimization problems having the following common property:Given a total task consisting of many subtasks,the problem asks to find a solution to complete only part of these subtasks.Examples include the k-Forest problem and the k-Multicut p
Location based services (LBS) are widely utilized,and determining the location of users\' IP is the foundation for LBS.Constrained by unstable delay and insufficient landmarks,the existing geolocation algorithms have problems such as low geolocation acc