Search

This paper presents a new method for measuring the cabin noise of a structure in a wind tunnel. A method for scaling the cabin sound was derived to obtain the cabin noise of a structure, and the derivation of the scaling procedure was based on a theoretical hypothesis regarding the cabin noise prediction for a scaled model in a wind tunnel. A frequency offset was generated because of the error introduced by model manufacture and installation, and a proposed modal test method was used to eliminate the frequency offset. Both a full-scale structure and scaled structure were measured in the wind tunnel tests. The cabin noise of the full-scale model was compared with the results obtained using the scaling procedure based on the scaled model. The comparisons of the measurement results indicate that the scaling procedures developed in this paper are effective for vibro-acoustic predictions in wind tunnels. Moreover, background noise tended to affect the results of the cabin sound for the wind tunnel test, and thus background noise should be prevented through specific design efforts. Keywords: Cabin noise, Scaling procedure, Turbulent boundary layer, Vibro-acoustic, Wind tunnel

The paper studies how common codes of artificial language in communication are developed in the laboratory. We find that codes emerging from an environment with more variable spatial positions tend to use a limited set of symbols to represent positions, whereas codes emerging from an environment with more variable geometric shapes tend to discriminate among shapes. The paper also experimentally shows that "language" affects the way its "speakers" share theview about a novel figure.

Social image tagging is becoming increasingly popular with the development of social website, where images are annotated with arbitrary keywords called tags. Most of present image tagging approaches are mainly based on the visual similarity or mapping between visual feature and tags. However, in the social media environment, images are always associated with multi-type of object information (i.e., visual content, tags, and user contact information) which makes this task more challenging. In this paper, we propose to fuse multi-type of information to tag social image. Specifically, we model social image tagging as a ''ranking and reinforcement'' problem, and a novel graph-based reinforcement algorithm for interrelated multi-type objects is proposed. When a user issue a tagging request for a query image, a candidate tag set is derived and a set of friends of the query user is selected. Then a graph which contains three types of objects (i.e., visual features of the query image, candidate tags, and friend users) is constructed, and each type of objects are initially ranked based on their weight and intra-relation. Finally, candidate tags are re-ranked by our graph-based reinforcement algorithm which takes into consideration both inter-relation with visual features and friend users, and the top ranked tags are saved. Experiments on real-life dataset demonstrate that our algorithm significantly outperforms state-of-the-art algorithms.

This paper experimentally explores how compositional grammars in artificial codes emerge and are sustained. In a pure coordination game with no conflict of interest, the sender sends a message that is an arbitrary string from available symbols with no prior meaning to indicate an abstract geometrical figure to the receiver. We find strong evidence from the laboratory for the emergence of compositional grammars in the subjects' common codes that facilitate learning efficiency. Moreover, when there is a scarcity of symbols in the repertoire, some groups in our experiments developed languages with positional compositionality, meaning the same symbol has different interpretations depending on its position in a string, whereas other groups developed language structures that are not compositional but still efficient in communication.

Online social network has become a popular way for users to express themselves, connect and share information with each other. However, in online social networks, the connections between different users are all in binary status, which neglects the relationship strengths between them. Meanwhile, the relationship strength between different users is activity field specific. In different activity fields, such as traveling, shopping, and sport, the relationship strengths between the same users may vary significantly. Therefore, in this paper we propose a general framework to measure the relationship strengths between different users, taking consideration not only the user's profile information but also the interaction activities and the activity fields. We conduct the experiments on Facebook dataset and the results show that the proposed framework is promising and can be used to improve the performances of various applications.

Video recommendation is an important tool to help people access interesting videos. In this paper, we propose a universal scheme to integrate rich information for personalized video recommendation. Our approach regards video recommendation as a ranking task. First, it generates multiple ranking lists by exploring different information sources. In particular, one novel source user's relationship strength is inferred through the online social network and applied to recommend videos. Second, based on multiple ranking lists, a multi-task rank aggregation approach is proposed to integrate these ranking lists to generate a final result for video recommendation. It is shown that our scheme is flexible that can easily incorporate other methods by adding their generated ranking lists into our multi-task rank aggregation approach. We conduct experiments on a large dataset with 76 users and more than 11,000 videos. The experimental results demonstrate the feasibility and effectiveness of our approach.

With internet delivery of video content surging to an un-precedented level, video recommendation has become an important approach for helping people access interesting videos. In this paper, we propose a novel approach to integrate viewing history for personalized video recommendation. For a given user, our approach calculates a recommendation score for each video candidate, that composes of two parts: the interest degree of this video by the user's friends, and the taste similarities between the user and his friends. We measure the interest degree of each video by considering its textual, visual and popularity information. Meanwhile, we construct tag set for each user based on his/her viewing history to estimate the taste similarities between different users. The final recommended videos are ranked according to the accumulated recommendation scores from different recommenders. We conduct experiments with 45 users and more than 11, 000 videos, and the results demonstrate the feasibility and effectiveness of our approach.

Video recommendation is a hot research topic to help people access interesting videos. The existing video recommendation approaches include CBF, CF and HF. However, these approaches treat the relationships between all users as equal and neglect an important fact that the acquaintances or friends may be a more reliable source than strangers to recommend interesting videos. Thus, in this paper we propose a novel approach to improve the accuracy of video recommendation. For a given user, our approach calculates a recommendation score for each video candidate that composes of two parts: the interest degree of this video by the user's friends, and the relationship strengths between the user and his friends. The final recommended videos are ranked according to the accumulated recommendation scores from different recommenders. We conducted experiments with 45 participants and the results demonstrated the feasibility and effectiveness of our approach.

Video recommendation is an important approach for helping people to access interesting videos. In this paper, we propose a scheme to integrate rich information for video recommendation. We regard video recommendation as a ranking problem and generate multiple ranking lists by exploring different information sources. A multi-task rank aggregation approach is proposed to integrate the ranking lists for different users in a joint manner. Our scheme is flexible and can easily incorporate other methods by adding their generated ranking lists into our multi-task learning algorithm. We conduct experiments with 76 users and more than 10,000 videos. The results demonstrate the feasibility and effectiveness of our approach.

Automatic video recommendation is involved in an attempt to tackle the information-overload problem, aiming to present the personalized video list to the user. This paper presents a novel approach to improve the accuracy of the video recommendation by combining the content-based filtering (CBF) method and the collaborative filtering (CF) method. Multimodal information is utilized to calculate the similarity among different videos to overcome the sparseness problem by CF method. We conduct experiments on a dataset of more than 11,000 videos and the results demonstrate the feasibility and effectiveness of our approach.

Community Question Answer (cQA) archives contain rich sources of knowledge on extensive topics, in which the quality of the submitted answer is uneven, ranging from excellent detailed answers to completely unrelated content. We propose a framework to generate complete, relevant, and trustful answer summaries. The framework discusses answer summarization in terms of maximum coverage problem with knapsack constraint on conceptual level. Global inference algorithm is employed to extract sentences according to the saliency scores of concepts. The saliency score of each concept is assigned through a two-layer graph-based random walk model incorporating the user social features and text content from answers. The experiments are implemented on a data set from Yahoo! Answer. The results show that our method generates satisfying summaries and is superior to the state-of-the-art approaches in performance.

A time series (or equivalently a data stream) consists of data arriving in time order. Single or multiple data streams arise in fields including physics, finance, medicine, and music, to name a few. Often the data comes from sensors (in physics and medicine for example) whose data rates continue to improve dramatically as sensor technology improves and as the number of sensors increases. So fast algorithms become ever more critical in order to distill knowledge from the data. This paper presents our recent work regarding the incremental computation of various primitives: windowed correlation, matching pursuit, sparse null space discovery and elastic burst detection. The incremental idea reflects the fact that recent data is more important than older data. Our StatStream system contains an implementation of these algorithms, permitting us to do empirical studies on both simulated and real data.

Financial time series streams are watched closely by millions of traders. What exactly do they look for and how can we help them do it faster? Physicists study the time series emerging from their sensors. The same question holds for them. Musicians produce time series. Consumers may want to compare them. This tutorial presents techniques and case studies for four problems:1. Finding sliding window correlations in financial, physical, and other applications.2. Discovering bursts in large sensor data of gamma rays.3. Matching hums to recorded music, even when people don't hum well.4. Maintaining and manipulating time-ordered data in a database setting.This tutorial draws mostly from the book High Performance Discovery in Time Series: techniques and case studies, Springer-Verlag 2004. You can find the power point slides for this tutorial at http://cs.nyu.edu/cs/faculty/shasha/papers/sigmod04.ppt.The tutorial is aimed at researchers in streams, data mining, and scientific computing. Its applications should interest anyone who works with scientists or financial "quants." The emphasis will be on recent results and open problems. This is a ripe area for further advance.

Autonomous robot architectures typically fall within two camps, top-down symbolic AI systems, and physically grounded bottom-up reactive systems. In this paper, we present a novel behavior fusion based architecture for the control of an intelligent powered wheelchair, which takes advantage of the modularity presented by the reactive bottom-up design, the inference mechanism of symbolic AI systems, and the flexibility of the layered architecture. Thus, this control architecture of wheelchair can realize the multi-level information fusion and facilitates task and information sharing and trading between man and machine. In addition, the basic behaviors and their fusion algorithm are also described.

In this paper ongoing work on an intelligent interactive wheelchair is presented. Also described is a new approach that integrates multimodal information for the navigational control of the wheelchair using a neuro-fuzzy network. Experiments show that the multimodal approach is capable of navigational control for the wheelchair, and of facilitating task and information sharing and trading between human and machine.

In this paper a novel approach of self-localization for an intelligent wheelchair is presented. Different from previous approaches we adopt a two-stage matching strategy to position control of an intelligent wheelchair, which is necessary if the current location is not known due to re-initialisation or an accident. Based on the landmark tree developed here, we combine a method similar to image-retrieval and a Bayes law based matching algorithm to locate the wheelchair in multiple resolutions, which is related to the requirement of task-completion. A new path-planning algorithm is introduced, which is originated from our landmark tree and is easy to use.