Reconstructing high-resolution (HR) 3D Gaussian Splatting (3DGS) models from low-resolution (LR) inputs remains challenging due to the lack of fine-grained textures and geometry. Existing methods typically rely on pre-trained 2D super-resolution (2DSR) models to enhance textures, but suffer from 3D Gaussian ambiguity arising from cross-view inconsistencies and domain gaps inherent in 2DSR models. We propose IE-SRGS, a novel 3DGS SR paradigm that addresses this issue by jointly leveraging the complementary strengths of external 2DSR priors and internal 3DGS features. Specifically, we use 2DSR and depth estimation models to generate HR images and depth maps as external knowledge, and employ multi-scale 3DGS models to produce cross-view consistent, domain-adaptive counterparts as internal knowledge. A mask-guided fusion strategy is introduced to integrate these two sources and synergistically exploit their complementary strengths, effectively guiding the 3D Gaussian optimization toward high-fidelity reconstruction. Extensive experiments on both synthetic and real-world benchmarks show that IE-SRGS consistently outperforms state-of-the-art methods in both quantitative accuracy and visual fidelity.

The development of Large AI Models (LAMs) for wireless communications, particularly for complex tasks like spectrum sensing, is critically dependent on the availability of vast, diverse, and realistic datasets. Addressing this need, this paper introduces the ChangShuoRadioData (CSRD) framework, an open-source, modular simulation platform designed for generating large-scale synthetic radio frequency (RF) data. CSRD simulates the end-to-end transmission and reception process, incorporating an extensive range of modulation schemes (100 types, including analog, digital, OFDM, and OTFS), configurable channel models featuring both statistical fading and site-specific ray tracing using OpenStreetMap data, and detailed modeling of realistic RF front-end impairments for various antenna configurations (SISO/MISO/MIMO). Using this framework, we characterize CSRD2025, a substantial dataset benchmark comprising over 25,000,000 frames (approx. 200TB), which is approximately 10,000 times larger than the widely used RML2018 dataset. CSRD2025 offers unprecedented signal diversity and complexity, specifically engineered to bridge the Sim2Real gap. Furthermore, we provide processing pipelines to convert IQ data into spectrograms annotated in COCO format, facilitating object detection approaches for time-frequency signal analysis. The dataset specification includes standardized 8:1:1 training, validation, and test splits (via frame indices) to ensure reproducible research. The CSRD framework is released at https://github.com/Singingkettle/ChangShuoRadioData1The dataset is designed to be fully reproducible using the provided framework, configurations, and configurable fixed random seeds to accelerate the advancement of AI-driven spectrum sensing and management.

Alzheimer’s Disease (AD) is a complex neurodegenerative disorder marked by memory loss, executive dysfunction, and personality changes. Early diagnosis is challenging due to subtle symptoms and varied presentations, often leading to misdiagnosis with traditional unimodal diagnostic methods due to their limited scope. This study introduces an advanced multimodal classification model that integrates clinical, cognitive, neuroimaging, and EEG data to enhance diagnostic accuracy. The model incorporates a feature tagger with a tabular data coding architecture and utilizes the TimesBlock module to capture intricate temporal patterns in Electroencephalograms (EEG) data. By employing Cross-modal Attention Aggregation module, the model effectively fuses Magnetic Resonance Imaging (MRI) spatial information with EEG temporal data, significantly improving the distinction between AD, Mild Cognitive Impairment, and Normal Cognition. Simultaneously, we have constructed the first AD classification dataset that includes three modalities: EEG, MRI, and tabular data. Our innovative approach aims to facilitate early diagnosis and intervention, potentially slowing the progression of AD. The source code and our private ADMC dataset are available at https://github.com/JustlfC03/MSTNet.

Automatic Modulation Classification (AMC) using deep learning techniques has become a prominent area of research, showcasing considerable practical applications. However, the present AMC deep learning network, trained on a specific channel model, performs poorly in a new channel scenario. To deal with this, an adversarial semi-supervised domain adaptation method is proposed. Specifically, classification accuracy, cluster sensitivity, and distribution distance are optimized together, utilizing a three-stage iterative training approach. As a result, the proposed model has achieved robust performance with a limited amount of labeled data when shifting to a new channel.

Jamming attack is a critical technique in communication countermeasures. This paper proposes a novel jamming method that employs Denoising Diffusion Proba-bilistic Models (DDPM) to generate distorted signals, which effectively increase the bit error rate (BER). The underlying mechanism functions similarly to a parroting technique, where the attacker replicates the transmission behavior during the communication process and transmits nonsensical information to cause interference. Compared to additive white Gaussian noise (AWGN) jamming, the proposed method demonstrates a more favorable energy efficiency ratio.

The power amplifier (PA) is an indispensable component in wireless communication systems, while the nonlinearity induced by PA can lead to significant performance degradation. The conventional nonlinearity equalization (NLE) method can effectively mitigate the nonlinear effects and provide superior BER performance but requires intensive computational complexity. To this end, we propose a novel NLE method in the time domain, which can significantly reduce the computational complexity without sacrificing the BER performance. Specifically, we first propose a novel PA representation of the sum of products (SPs), which is a unified time-domain representation for several typical memory and memoryless PA models. On this basis, the SPs-iterative least square equalizer (SPs-ILSE) method is proposed to mitigate the impact of both the memory and memoryless PA’s nonlinear distortions at the receiver side. The computational complexity of complex multiplication (CCCM) in the proposed method isO(NlogN) for the memoryless PA models andO(KN2) for the memory PA models. Moreover, considering the commonly utilized PA models, we also derive the closed-form expression for the achievable SINR of the SPs-ILSE method in the ideal conditions. Numerical results show that (i) the closed-form SINR expression is valid for both the memory and memoryless scenarios (ii) the proposed method exhibits the superior bit error rate (BER) performance in comparison to several relevant nonlinear signal processing methods such as digital pre-distortion (DPD), and power amplifier nonlinearity cancellation (PANC) (iii) the proposed NLE method achieves the same BER performance as the previous NLE method, i.e., reconstruction of distorted signals (RODS), while the CCCM of the proposed method is much lower.

Automatic modulation classification (AMC) is a crucial technique for the design of intelligent transceivers and has received considerable research attention. Conventional feature-based (FB) methods have the advantage of low computational complexity. However, these methods are highly sensitive to the distribution shifts of the received signal caused by the variation of channel effects and have rarely been studied in orthogonal frequency division multiplexing (OFDM) systems under unseen synthetic channels with multipath fading effects, carrier frequency offset (CFO), phase offset (PO) and additive noise. To solve this problem, this paper proposes a novel FB method using the error vector magnitude (EVM) features for AMC tasks (termed as EVM-AMC), which can achieve reliable classification performance for the communication scenarios considering unseen synthetic channels in OFDM systems. Specifically, we first propose the axisymmetric mapping-based self-circulant differential division (AM-SCDD) algorithm to convert the received signal into the non-negative spectral quotient (NNSQ) sequence, deeply suppressing the synthetic channel effects. Subsequently, we derive the EVM features by analyzing the matched error vectors between the generated NNSQ sequence and the predefined NNSQ constellation symbol (NNSQCS) masks. During this process, a percentile-based filter is utilized to remove the outliers in each matched error vector. Finally, the feature samples collected from various channel conditions are sent to the multi-class support vector machine (SVM) classifiers for training and testing. Two candidate modulation type sets are employed to evaluate the performance of the proposed EVM-AMC method under both the constant and changing channel conditions. Our numerical results demonstrate that 1) the proposed method exhibits impressive robustness and generalization when dealing with unseen synthetic channels, 2) the proposed method yields the best classification performance when compared to the conventional FB AMC methods in the presence of channel effects.

Signal detection and modulation classification are two crucial tasks in various wireless communication systems. Different from prior works that investigate them independently, this paper studies the joint signal detection and automatic modulation classification (AMC) by considering a realistic and complex scenario, in which multiple signals with different modulation schemes coexist at different carrier frequencies. We first generate a coexisting RADIOML dataset (CRML23) to facilitate the joint design. Different from the publicly available AMC dataset, ignoring the signal detection step and containing only one signal, our synthetic dataset covers the more realistic multiple-signal coexisting scenario. Then, we present a joint framework for detection and classification (JDM) for such a multiple-signal coexisting environment, which consists of two modules for signal detection and AMC, respectively. In particular, these two modules are interconnected using a designated data structure called “proposal”. Finally, we conduct extensive simulations over the newly developed dataset, which demonstrate the effectiveness of our designs. Our code and dataset are now available as open-source resources athttps://github.com/Singingkettle/ChangShuoRadioData.

Sequential recommenders have been widely used in industry due to their strength in modeling user preferences. While these models excel at learning a user’s positive interests, less attention has been paid to learning from negative user feedback. Negative user feedback is an important lever of user control, and comes with an expectation that recommenders should respond quickly and reduce similar recommendations to the user. However, negative feedback signals are often ignored in the training objective of sequential retrieval models, which primarily aim at predicting positive user interactions. In this work, we incorporate explicit and implicit negative user feedback into the training objective of sequential recommenders in the retrieval stage using a "not-to-recommend" loss function that optimizes for the log-likelihood of not recommending items with negative feedback. We demonstrate the effectiveness of this approach using live experiments on a large-scale industrial recommender system. Furthermore, we address a challenge in measuring recommender responsiveness to negative feedback by developing a counterfactual simulation framework to compare recommender responses between different user actions, showing improved responsiveness from the modeling change.

Radio frequency fingerprint identification (RFFI) is a promising physical layer security technique that employs the hardware-introduced features extracted from the received signals for device identification. In this paper, we consider an RFFI problem in the presence of hybrid time-varying distortions (HTVDs) induced by multipath fading channel, carrier frequency offset (CFO), and phase offset. To solve this problem, an HTVDs-robust RFFI framework is proposed. Firstly, we derive that the residual HTVDs after CFO correction can be approximated as multiplicative interference in the frequency domain. Secondly, we define a novel signal analysis dimension named spectral quotient (SQ) representation and then present the spectral circular shift division (SCSD) method to generate the HTVDs-robust SQ signals, where the multiplicative interference can be suppressed. Thereafter, the statistics including root mean square (RMS), variance (VAR), skewness (SKE), and kurtosis (KUR) are extracted from the real and imaginary components of the SQ signals, respectively. Finally, the statistical features are used for the training and testing of the support vector machine (SVM) classifiers. To further enhance the performance of the proposed RFFI scheme, we also present the spectral circular multi-shift division (SCMSD) method, which increases the flexibility in the generation of the HTVDs-robust SQ signals. Given what we knew, this is the first time attempting to mitigate the HTVDs by leveraging the strong frequency correlation at the neighboring subcarriers in the multivariate hypothesis tasks. Compared to several handcraft feature-based RFFI methods, the proposed method exhibits superior identification accuracy and strong robustness. Experimental results show that the proposed RFFI scheme can achieve the accuracy of 91.3%with five devices and 86.4% with sixteen devices when the classifiers are trained with the additive white Gaussian noise but are tested with the Rayleigh channel.

Automatic modulation classification (AMC) is to identify the modulation type of a received signal, which plays a vital role to ensure the physical-layer security for Internet of Things (IoT) networks. Inspired by the great success of deep learning in pattern recognition, the convolutional neural network (CNN) and recurrent neural network (RNN) are introduced into the AMC. In general, there are two popular data formats used by AMC, which are the in-phase/quadrature (I/Q) representation and amplitude/phase (A/P) representation, respectively. However, most of AMC algorithms aim at structure innovations, while the differences and characteristics of I/Q and A/P are ignored to analyze. In this article, lots of popular AMC algorithms are reproduced and evaluated on the same data set, where the I/Q and A/P are used, respectively, for comparison. Based on the experimental results, it is found that: 1) CNN-RNN-like algorithms using A/P as input data are superior to those using I/Q at high signal-to-noise ratio (SNR), while it has an opposite result in low SNR and 2) the features extracted from I/Q and A/P are complementary to each other. Motivated by the aforementioned findings, a multitask learning-based deep neural network (MLDNN) is proposed, which effectively fuses I/Q and A/P. In addition, the MLDNN also has a novel backbone, which is made up of three blocks to extract discriminative features, and they are CNN block, bidirectional gated recurrent unit (BiGRU) block, and a step attention fusion network (SAFN) block. Different from most of CNN-RNN-like algorithms (i.e., they only use the last step outputs of RNN), all step outputs of BiGRU can be effectively utilized by MLDNN with the help of SAFN. Extensive simulations are conducted to verify that the proposed MLDNN achieves superior performance in the public benchmark.

Three-dimensional detection is essential for autonomous driving and intelligent transportation system, as it enables vehicles to detect and track surrounding objects. Usually, autonomous vehicles are equipped with multiple sensing modalities to achieve robust and precise detection. This work focuses on fusing millimeter-wave radar data with monocular images, as radar can make up for the lack of explicit depth information. We propose a novel approach that fuses radar data and images at the feature level for 3-D detection. Radar points are first merged into a raw feature map with data set statistics by a novel transformation method. With this transformation, radar features can be extracted by convolutional neural networks and fused with image features. Object properties, including location, dimension, and rotation are regressed from the fused features. In this article, the proposed fusion strategy is implemented with a keypoint-based 3-D detection framework and evaluated on the challenging NuScenes data set. Experimental results suggest that the fusion of radar data promotes 3-D detection capability in public benchmarking.

The Internet of Things (IoT) permeates every aspect of our daily lives as billions of interconnected devices are deployed in the physical world. However, IoT networks operate in an untrusted environment and often suffer from many malicious active attacks. Automatic modulation classification (AMC), which can identify the modulation format of intercepted signals without prior knowledge, is a vital technology in countering physical-layer threats of IoT. However, most of the existing algorithms assume the channel is time invariant, and the AMC in time-varying channels is not been well studied. To deal with this dilemma, a novel AMC algorithm MCBLDN consisting of multiple convolutional neural networks (CNNs), a bidirectional long short-term memory network (BLSTM), and a deep neural network (DNN) is proposed. In MCBLDN, a multislot constellation diagram (CD) method is proposed to extract time-evolution characteristics for generating more discriminative features. Specifically, different grayscale subimages generated by slotted CDs are processed serially by their respective CNNs. Therefore, MCBLDN is overparameterized and time consuming. In addition, the frequency offset and phase offset caused by time-varying channels are neglected in MCBLDN, which is detrimental to the performance of AMC. To address the mentioned disadvantages, a lightweight MCBLDN with a spatial transformer network (SLCBDN) is proposed. First, the multiple CNNs in MCBLDN are pruned into a lightweight classification model, and the input data are rearranged to facilitate parallel processing by the lightweight CNN. Additionally, the spatial transformer network (STN) is utilized to reduce the influence of frequency offset and phase offset. Numerical results verify that the proposed method achieves superior performance and higher speed compared to the baseline algorithm MCBLDN.

For future connected automated vehicles (CAVs) networks, the joint optimization of communication, sensing, and computing resources is crucial to guarantee the performance of cooperative automated driving’s safety, which is attracting more and more attention. However, the existing works have not considered the low-latency requirement for the raw perception data sharing with both wireless communication link capability and computing efficiency constraints, causing a serious threat to the cooperative automated driving’s safety in CAVs networks. In this article, a vehicle–road–base station cooperation architecture is designed, and a federated reinforcement learning (FRL)-based task offloading and resource allocation algorithm in the CAVs network is proposed to reduce the task execution delay with different communication and computing constraints. The problem of execution delay minimization is theoretically formulated and analyzed under three task practical offloading modes. To adapt to the dynamic topology of the CAVs network, we design a deep reinforcement learning algorithm to achieve the optimal task offloading and resource allocation. To further reduce the data transmission overhead of the centralized reinforcement learning algorithm, the FRL-enabled algorithm is proposed to minimize the execution delay of the optimal task offloading and resource allocation among multiple CAVs. Both the simulation and hardware testbed results verify that the proposed algorithms can not only reduce the execution delay and the communication overhead but also improve the system throughput.

Automatic modulation classification (AMC) identifies a received signal’s modulation scheme without prior knowledge of the intercepted signal, which enables significant applications in both the military and civilian domains. Inspired by the great success of deep learning (DL), lots of neural networks are introduced into AMC. To further improve classification performance, various complementary cues including in-phase/quadrature (I/Q), amplitude/phase (A/P), constellation, and other formats are used together to enhance the discrimination of the DL model, where only outputs of the last layer are used. In this paper, we find that different layers’ outputs in the DL model are also complementary to each other. As a result, a hierarchical classification head based convolutional gated deep neural network (HCGDNN) is proposed by utilizing different layers’ output, which only uses the I/Q cue. The proposed HCGDNN consists of three groups of convolutional neural networks (CNN) blocks, two groups of bidirectional gated recurrent units (BiGRU), and a hierarchical classification head. Compared to the long short-term memory (LSTM), the BiGRU has a smaller computational complexity and also releases the gradient dispersion and explosion in the training phase. With the help of the hierarchical classification head, three groups of modulation predictions are made for a received I/Q signal. After that, a novel nonlinear optimization fusion method is derived to generate fusion weights to fuse different groups, then a final classification decision is made. Compared to AMC methods using various cues, the proposed HCGDNN only uses I/Q cue and has low computational overhead. Numerical results suggest that the newly developed HCGDNN achieves superior performance on the public benchmark.To help other researchers, the source code will be uploaded to the github as long as the paper is published.

To address the issue of multi-scale detection, current detectors usually generate hierarchical feature pyramid by a naive combination of top-down features with lateral features. Due to the limited effective receptive fields in the top-down pathway, the generated regions are only associated with the neighbor regions of the coarse feature maps, which is harmful for pyramidal feature generation. And considering the weak representation of simply merging of top-down and lateral pathways, the pyramidal feature maps with strong semantics are difficult to obtain. In this paper, we present the Residual Dilation based Feature Pyramid Network (RDFPN) to exploit the inherent correlation of regions in feature pyramid. The goal of RDFPN is to produce more appropriate hierarchical feature maps for multi-scale detection. With Residual-50 network in Faster R-CNN framework, RDFPN outperforms the original FPN on the challenging COCO dataset without bells and whistles.

The ACM RecSys'18 Late-Breaking Results track (previously known as the Poster track) is part of the main program of the 2018 ACM Conference on Recommender Systems in Vancouver, Canada. The track attracted 48 submissions this year out of which 18 papers could be accepted resulting in an acceptance rated of 37.5%.

The technical barriers for conversing with recommender systems using natural language are vanishing. Already, there are commercial systems that facilitate interactions with an AI agent. For instance, it is possible to say "what should I watch" to an Apple TV remote to get recommendations. In this research, we investigate how users initially interact with a new natural language recommender to deepen our understanding of the range of inputs that these technologies can expect. We deploy a natural language interface to a recommender system, we observe users' first interactions and follow-up queries, and we measure the differences between speaking- and typing-based interfaces. We employ qualitative methods to derive a categorization of users' first queries (objective, subjective, and navigation) and follow-up queries (refine, reformulate, start over). We employ quantitative methods to determine the differences between speech and text, finding that speech inputs are typically longer and more conversational.

A relatively new feature in Google Play Store presents mobile app search results grouped by topic, helping users to quickly navigate and explore. The underlying Search Results Clustering (SRC) system faces several challenges, including grouping search results in topical coherent clusters as well as finding the appropriate level of granularity for clustering. We present AppGrouper, an alternative approach to algorithmic-only solutions, incorporating human input in a knowledge-graph-based clustering process. AppGrouper provides an interactive interface that lets domain experts steer the clustering process in early, mid, and late stages. We deployed and evaluated AppGrouper with internal experts. We found that AppGroup improved quality of algorithm-generated app clusters on 56 out of 82 search queries. We also found that the internal experts made more changes in early and mid stages for lower-quality algorithmic results, focusing more on narrow queries. Our result suggests, in some contexts, machine learning systems can greatly benefit from steering from human experts, creating a symbiotic working relationship.

Explanations are important for users to make decisions on whether to take recommendations. However, algorithm generated explanations can be overly simplistic and unconvincing. We believe that humans can overcome these limitations. Inspired by how people explain word-of-mouth recommendations, we designed a process, combining crowdsourcing and computation, that generates personalized natural language explanations. We modeled key topical aspects of movies, asked crowdworkers to write explanations based on quotes from online movie reviews, and personalized the explanations presented to users based on their rating history. We evaluated the explanations by surveying 220 MovieLens users, finding that compared to personalized tag-based explanations, natural language explanations: 1) contain a more appropriate amount of information, 2) earn more trust from users, and 3) make users more satisfied. This paper contributes to the research literature by describing a scalable process for generating high quality and personalized natural language explanations, improving on state-of-the-art content-based explanations, and showing the feasibility and advantages of approaches that combine human wisdom with algorithmic processes.

As users browse a recommender system, they systematically consider or skip over much of the displayed content. It seems obvious that these eye gaze patterns contain a rich signal concerning these users' preferences. However, because eye tracking data is not available to most recommender systems, these signals are not widely incorporated into personalization models. In this work, we show that it is possible to predict gaze by combining easily-collected user browsing data with eye tracking data from a small number of users in a grid-based recommender interface. Our technique is able to leverage a small amount of eye tracking data to infer gaze patterns for other users. We evaluate our prediction models in MovieLens -- an online movie recommender system. Our results show that incorporating eye tracking data from a small number of users significantly boosts accuracy as compared with only using browsing data, even though the eye-tracked users are different from the testing users (e.g. AUC=0.823 vs. 0.693 in predicting whether a user will fixate on an item). We also demonstrate that Hidden Markov Models (HMMs) can be applied in this setting; they are better than linear models in predicting fixation probability and capturing the interface regularity through Bayesian inference (AUC=0.823 vs. 0.757).

Pedestrian detection systems are receiving increasing attention in both industry and academia with the rapid development of autonomous automobiles which employ artificial intelligence. These systems must detect specific classes of objects such as pedestrians rather than generic objects. In this paper, we present a faster RCNN based pedestrian detection system which improves upon previous solutions. The proposed model takes arbitrary size images as inputs and generates bounding boxes and confidence scores for pedestrians. The system achieves good performance and is faster than the well known and frequently used methods in the literature.

To achieve high quality initial personalization, recommender systems must provide an efficient and effective process for new users to express their preferences. We propose that this goal is best served not by the classical method where users begin by expressing preferences for individual items - this process is an inefficient way to convert a user's effort into improved personalization. Rather, we propose that new users can begin by expressing their preferences for groups of items. We test this idea by designing and evaluating an interactive process where users express preferences across groups of items that are automatically generated by clustering algorithms. We contribute a strategy for recommending items based on these preferences that is generalizable to any collaborative filtering-based system. We evaluate our process with both offline simulation methods and an online user experiment. We find that, as compared with a baseline rate-15-items interface, (a) users are able to complete the preference elicitation process in less than half the time, and (b) users are more satisfied with the resulting recommended items. Our evaluation reveals several advantages and other trade-offs involved in moving from item-based preference elicitation to group-based preference elicitation.

Pinterest is a popular social networking site that lets people discover, collect, and share pictures of items from the Web. Among popular social media sites, Pinterest has by far the most skewed gender distribution: women are four times more likely than men to use it. To better understand this, we examined two factors that generally affect whether people try a social site and whether they continue using it: the external perception of a site (e.g., as conveyed in popular media) and the site's initial user experience. For the latter, we focused on the role of social bootstrapping, importing contacts from one social site to another. We conducted a survey study, finding that: perceptions of Pinterest among users and non-users of the site differed significantly; trying Pinterest led to substantial changes in user perceptions of the site; social bootstrapping affected users' initial impression of Pinterest, generally improving it for women and harming it for men. We present implications of our findings for design and research.

The essence of a recommender system is that it can recommend items personalized to the preferences of an individual user. But typically users are given no explicit control over this personalization, and are instead left guessing about how their actions affect the resulting recommendations. We hypothesize that any recommender algorithm will better fit some users' expectations than others, leaving opportunities for improvement. To address this challenge, we study a recommender that puts some control in the hands of users. Specifically, we build and evaluate a system that incorporates user-tuned popularity and recency modifiers, allowing users to express concepts like "show more popular items". We find that users who are given these controls evaluate the resulting recommendations much more positively. Further, we find that users diverge in their preferred settings, confirming the importance of giving control to users.

Pinterest is a popular social curation site where people collect, organize, and share pictures of items. We studied a fundamental issue for such sites: what patterns of activity attract attention (audience and content reposting)-- We organized our studies around two key factors: the extent to which users specialize in particular topics, and homophily among users. We also considered the existence of differences between female and male users. We found: (a) women and men differed in the types of content they collected and the degree to which they specialized; male Pinterest users were not particularly interested in stereotypically male topics; (b) sharing diverse types of content increases your following, but only up to a certain point; (c) homophily drives repinning: people repin content from other users who share their interests; homophily also affects following, but to a lesser extent. Our findings suggest strategies both for users (e.g., strategies to attract an audience) and maintainers (e.g., content recommendation methods) of social curation sites.

Community Question Answering (CQA) service enables its users to exchange knowledge in the form of questions and answers. By allowing the users to contribute knowledge, CQA not only satisfies the question askers but also provides valuable references to other users with similar queries. Due to a large volume of questions, not all questions get fully answered. As a result, it can be useful to route a question to a potential answerer. In this paper, we present a question routing scheme which takes into account the answering, commenting and voting propensities of the users. Unlike prior work which focuses on routing a question to the most desirable expert, we focus on routing it to a group of users - who would be willing to collaborate and provide useful answers to that question. Through empirical evidence, we show that more answers and comments are desirable for improving the lasting value of a question-answer thread. As a result, our focus is on routing a question to a team of compatible users. We propose a recommendation model that takes into account the compatibility, topical expertise and availability of the users. Our experiments over a large real-world dataset shows the effectiveness of our approach over several baseline models.

Over the past decade, social network sites have become ubiquitous places for people to maintain relationships, as well as loci of intense research interest. Recently, a new site has exploded into prominence: Pinterest became the fastest social network to reach 10M users, growing 4000% in 2011 alone. While many Pinterest articles have appeared in the popular press, there has been little scholarly work so far. In this paper, we use a quantitative approach to study three research questions about the site. What drives activity on Pinterest? What role does gender play in the site's social connections? And finally, what distinguishes Pinterest from existing networks, in particular Twitter? In short, we find that being female means more repins, but fewer followers, and that four verbs set Pinterest apart from Twitter: use, look, want and need. This work serves as an early snapshot of Pinterest that later work can leverage.