Search

Structural damping ratio which quantifies the energy dissipation of civil structures under external excitations plays a critical role in the seismic design and assessment of civil structures. In existing building design provisions and guidelines, however, the structural damping ratio is only suggested either as a single fixed value or as an optional value for the general structure type adopted. For example, damping ratio 5% is commonly recommended for all reinforced concrete (RC) structures in practical seismic design, which may not be sufficient to represent the realistic damping features of different RC structures under ground motions with different amplitudes. This research explored deeper understandings on the structural damping features of different RC structures under actual ground motion excitations. A series of seismic response records of RC structures were collected from the “Center for Engineering Strong Motion Data” (CESMD) database. These records were then categorized into three typical lateral resisting systems: moment-resisting frame systems, shear wall systems, and moment-resisting frame plus shear wall systems. The equivalent structural damping ratios for different systems of RC structures were then estimated based on the categorized response records with different amplitudes. Finally, an empirical statistical relationship was established, offering a refined basis for civil engineers to reasonably choose the equivalent damping ratios during the design and post-earthquake assessment of the RC structures.

Attention-based deep learning network models have shown obvious advantages on the sentence representation in many NLP tasks. While in the answer selection domain, applying attention-based deep learning model to capture complex semantic relations between question and answer is an extremely challenging task. In this paper, instead of simply using max-pooling in the pooling layer, we propose the two-level attentive pooling model which can efficiently select several key and high semantic-related matching words in the question-answer pair to improve the accuracy of answer selection. Specially, our model is built on top of the hybrid network which includes GRU and CNN to encode the complex sentence representation. The experimental evaluation on two popular datasets shows that our model has the good effectiveness and achieves the state-of-art performance in the answer selection task.

The Spiral of Silence has been studied widely in the traditional propagation field. However, to our best knowledge, no one has clearly verified the Spiral of Silence in social networks based on the real information diffusion data. In this paper, four factors including width, depth, message sentiment, and modularity of information diffusion trees are analyzed to verify the applicability of the theory. Disparities between majority and minority are found to different extents in various topics. Based on Spiral of Silence, polarity prediction of users’ review without considering the semantic meaning of content is proposed and discovered. The results indicate that opinions of people in propagation are impacted by social environment and their friends. The Anti-Spiral of Silence, an extension of Spiral of Silence, has been found to play a significant role in leading rational public opinion and revealing truth in social networks. Our works of both Spiral of Silence and Anti-Spiral of Silence will enrich research results on the study and application of propagation effects.

With the explosive growth of social network, the exploitation of social information in recommendation models has become increasingly significant. However, most existing models only made use of users’ social information, ignoring the value of items’ social information. Based on the above fact, we present a listwise learning to rank recommendation model, UIContextListRank, which generates a ranked list of items for individual users directly. We employ matrix factorization to construct a listwise objective function that measures the difference between the predicted lists and the real ones. Furthermore, we express users’ social contextual information as their trust friends and items’ social contextual information as their concurrent items, and incorporate the social contextual information of both users’ and items’ into the listwise model to improve recommendation quality. Moreover, we implement our proposed model in a distributed environment to tackle the challenge of overwhelming data. Experiments have been conducted on two real-world datasets to evaluate the proposed model. And the experimental results prove the model’s effectiveness and efficiency.

In this paper, we propose a real-time recommender system (PRACE) for taxi drivers to find a next passenger and start a new trip efficiently, based on historical GPS trajectories of taxis. To provide high-quality passenger-seeking advice, PRACE takes passenger prediction, road condition estimation, and earnings into ranking simultaneously. Different from many previous researchers, we not only pay more attention to the driving context of taxis (i.e., driving directions, positions, etc.) but also extract meaningful representations of these attributes, using deep neural networks. To enhance the effect of learning, the result of statistics is added to the input of models. Relying on the map meshing method, we treat the prediction task as a multi-classification problem rather than a regression problem and make comparisons with several state-of-the-art methods. Finally, we evaluate our method through extensive experiments, using GPS trajectories generated by more than 10,000 taxis from the same company over a period of two months. The results verify the effectiveness, efficiency, and availability of our recommender system.

Finding the key properties of connectivity in Vehicle Ad-hoc Network (VANET) is an important challenge because of a wide geographic range, an uneven distribution of vehicles, and low coupling of interconnections. The prior work has mostly concentrated on VANET which are mainly through a WAVE wireless network protocol to implement a hop-by-hop inter-vehicle (V2V) communication. It has a low degree of verisimilitude and lacks formal analysis and theoretical methods to deal with a large-scale open network environment. In this paper, we give some important results on the key properties of connectivity in VANET: (1) The number of edges and nodes obey the Densification Power Law. (2) An entire VANET is not connected. (3) Dense vehicle community contains both vehicles with large degree and small ones. (4) The neighbors’ connection of a vehicle with a large degree is sparse. This work should motivate VANET researchers, practitioners, and new comers to know the nature of key properties of connectivity in VANET.

Southeast Asian cities such as Jakarta (Indonesia) present traffic patterns that differ importantly from ‘Western’ cities. This includes the typology of vehicles, lane utilization and distinctions between pedestrian and vehicular routes. Such differences are maintained, and sometimes accentuated, during a sudden evacuation, such as the one that might occur as consequence of a (actual or perceived) tsunami hazard. Most of the commonly used traffic models are not designed for reproducing these patterns, assuming that the vehicles would follow queuing patterns or lane divisions, and allowing a limited interaction between pedestrians and vehicles. In reality, during a non-staged sudden evacuation, the portion of the road occupied by pedestrians can vary significantly, seemingly depending on the ratio between pedestrians and vehicles. This paper presents the results of a study aiming to develop a hybrid evacuation model able to target these issues. In particular, the model aims to describe the vehicular traffic consistently with the local traffic patterns, and represent a strong interrelationship between vehicular and pedestrian traffic, under the constraint of modelling a large number of evacuees. Results from preliminary applications are presented and compared with existing models.