Daqing Zhang

Bio: Daqing Zhang is an academic researcher from Peking University. The author has contributed to research in topics: Context (language use) & Mobile computing. The author has an hindex of 67, co-authored 331 publications receiving 16675 citations. Previous affiliations of Daqing Zhang include Institut Mines-Télécom & Institute for Infocomm Research Singapore.

Correlating MSM's Mental Health with Usage Behaviors on MSM-Specific Social Applications

Abstract: Homophobia and discrimination towards men who have sex with men (MSM) make the mental health of this community a severe concern. The prevalence of mobile social apps for MSM provides a new channel to study their mental health issues. However, the correlation between the psychological states of MSM and their behaviors on social apps are still uninvestigated. In this paper, we conduct a case study of 103 MSM in China to explore whether the User Generated Content, Profile, Mobility information of MSM on social apps correlate with their depression, loneliness, anxiety and stress level. The analysis results indicate that the mental health disorder level of MSM is correlated with the emotion expressed in posts, whether the user has filled in self-description blank, the social relationship among online friends, frequency of visiting popular gays-meeting places. Our findings imply future opportunities in early psychological problems detection and intervention.

Interlayer directional coupling thermo-optic waveguide switches based on functionalized epoxy-crosslinking polymers.

Abstract: In this study, interlayer directional coupling (DC) thermo-optic (TO) waveguide switches were designed and fabricated using functionalized epoxy-crosslinking polymers. Fluorinated SU-8 (FSU-8) with a photo-initiating epoxy-crosslinking network was self-synthesized as a waveguide core material. A copolymer of methyl methacrylate and glycidyl methacrylate P(MMA-co-GMA) with a thermo-initiating epoxy crosslinking structure was self-synthesized as a waveguide cladding material. Compared with commercial pure SU-8 and PMMA, FSU-8 exhibited a lower absorption loss and P(MMA-co-GMA) exhibited a higher thermal stability. Using epoxy-crosslinking functionalized polymers, the structure of the waveguides and electrode heaters were optimized, and the performance parameters of the interlayer DC TO switches were simulated. At a signal wavelength of 1550 nm, the insertion loss, extinction ratio, and power consumption of the actual interlayer devices were measured as 6.7 dB, 15.6 dB, and 9 mW, respectively. The rising and falling response times of the TO switches were obtained as 631.6 µs and 362 µs, respectively. The self-leveling ability and solvent resistance characteristic of the epoxy-crosslinking network for FSU-8 and P(MMA-co-GMA) may guarantee the realization of interlayer DC TO waveguide switches. The proposed technique will be suitable for photonic integrated waveguide chips with multilayer stacking dynamic optical information interactions.

The first international symposium on social and community intelligence (SCI'11)

Abstract: Social and Community Intelligence (SCI) represents an emerging area that aims at revealing individual/group behaviors, social interactions as well as community dynamics by mining the digital traces left by people while interacting with cyber-physical spaces. The digital traces are generated mainly from three information sources: Internet and Web applications, static infrastructure, mobile devices and wearable sensors. In this workshop we hope to get people from different disciplines together to share their visions and insights on how to tackle the challenges faced by SCI, such as participatory sensing, heterogeneous data fusion, intelligence extraction, privacy issues, and so on.

Ultra-Broadband and Compact TM-Pass Polarizer Based on Graphene-Buried Polymer Waveguide

Abstract: We report an ultra-broadband and compact TM-pass polarizer based on graphene-buried polymer waveguides. The characteristic parameters of the polarizer were carefully designed and optimized. The standard microfabrication processes were employed to fabricate the device. The presented polarizers exhibit high polarization-dependent transmission imposing a TE mode cutoff while leaving the TM mode almost unaffected. We experimentally demonstrated the polarizer that has an ultra-high extinction ratio of more than 22.9 dB and 41.9 dB for the monolayer graphene film placed on the surface of core layer and buried in the center of core layer, respectively, and as low insertion loss as ~4.0 dB for the TM mode with the bandwidth over 110 nm. The presented polarizer has the advantages of high extinction ratio, ultra-broadband, low cost, and easy integration with other polymer-based planar lightwave devices.

Supporting development of context-aware applications using semantic space toolkit

Abstract: In order to facilitate rapid development of context-aware applications, there is a need for architectural support in the entire context processing flow, and improved programming abstractions that ease the prototyping. In this paper, a toolkit called Semantic Space, is proposed to support rapid prototyping of context-aware applications via a set of programming abstractions on context processing. The functionality encapsulated in the toolkit handles the common, time-consuming and low-level details in context acquisition, aggregation, storage and inference. Architectural design and implementation issues of the Semantic Space toolkit are discussed in detail. Finally, a case study on building a mobile situation aware phone is described to illustrate the validity of our approach and usability of the toolkit.

Context Aware Computing for The Internet of Things: A Survey

Abstract: As we are moving towards the Internet of Things (IoT), the number of sensors deployed around the world is growing at a rapid pace. Market research has shown a significant growth of sensor deployments over the past decade and has predicted a significant increment of the growth rate in the future. These sensors continuously generate enormous amounts of data. However, in order to add value to raw sensor data we need to understand it. Collection, modelling, reasoning, and distribution of context in relation to sensor data plays critical role in this challenge. Context-aware computing has proven to be successful in understanding sensor data. In this paper, we survey context awareness from an IoT perspective. We present the necessary background by introducing the IoT paradigm and context-aware fundamentals at the beginning. Then we provide an in-depth analysis of context life cycle. We evaluate a subset of projects (50) which represent the majority of research and commercial solutions proposed in the field of context-aware computing conducted over the last decade (2001-2011) based on our own taxonomy. Finally, based on our evaluation, we highlight the lessons to be learnt from the past and some possible directions for future research. The survey addresses a broad range of techniques, methods, models, functionalities, systems, applications, and middleware solutions related to context awareness and IoT. Our goal is not only to analyse, compare and consolidate past research work but also to appreciate their findings and discuss their applicability towards the IoT.

Membership Inference Attacks Against Machine Learning Models

Abstract: We quantitatively investigate how machine learning models leak information about the individual data records on which they were trained. We focus on the basic membership inference attack: given a data record and black-box access to a model, determine if the record was in the model's training dataset. To perform membership inference against a target model, we make adversarial use of machine learning and train our own inference model to recognize differences in the target model's predictions on the inputs that it trained on versus the inputs that it did not train on. We empirically evaluate our inference techniques on classification models trained by commercial "machine learning as a service" providers such as Google and Amazon. Using realistic datasets and classification tasks, including a hospital discharge dataset whose membership is sensitive from the privacy perspective, we show that these models can be vulnerable to membership inference attacks. We then investigate the factors that influence this leakage and evaluate mitigation strategies.