Networks with node covariates are commonplace: for example, people in a social network have interests, or product preferences, etc. If we know the covariates for some nodes, can we infer them for the remaining nodes? In this paper we propose a new similarity measure between two nodes based on the patterns of their 2-hop neighborhoods. We show that a simple algorithm (CN-VEC) like nearest neighbor regression with this metric is consistent for a wide range of models when the degree grows faster than $n^{1/3}$ up-to logarithmic factors, where $n$ is the number of nodes. For "low-rank" latent variable models, the natural contender will be to estimate the latent variables using SVD and use them for non-parametric regression. While we show consistency of this method under less stringent sparsity conditions, our experimental results suggest that the simple local CN-VEC method either outperforms the global SVD-RBF method, or has comparable performance for low rank models. We also present simulated and real data experiments to show the effectiveness of our algorithms compared to the state of the art.

Factorization Machines (FM) have been widely applied in industrial applications for recommendations. Traditionally FM models are trained in batch mode, which entails training the model with large datasets every few hours or days. Such training procedure cannot capture the trends evolving in real time with large volume of streaming data. In this paper, we propose an online training scheme for FM with the alternating least squares (ALS) technique, which has comparable performance with existing batch training algorithms. We incorporate an online update mechanism to the model parameters at the cost of storing a small cache. The mechanism also stabilizes the training error more than a traditional online training technique like stochastic gradient descent (SGD) as data points come in, which is crucial for real-time applications. Experiments on large scale datasets validate the efficiency and robustness of our method.

People belong to multiple communities, words belong to multiple topics, and books cover multiple genres; overlapping clusters are commonplace. Many existing overlapping clustering methods model each person (or word, or book) as a non-negative weighted combination of "exemplars" who belong solely to one community, with some small noise. Geometrically, each person is a point on a cone whose corners are these exemplars. This basic form encompasses the widely used Mixed Membership Stochastic Blockmodel of networks and its degree-corrected variants, as well as topic models such as LDA. We show that a simple one-class SVM yields provably consistent parameter inference for all such models, and scales to large datasets. Experimental results on several simulated and real datasets show our algorithm (called SVM-cone) is both accurate and scalable.

The problem of finding overlapping communities in networks has gained much attention recently. Optimization-based approaches use non-negative matrix factorization (NMF) or variants, but the global optimum cannot be provably attained in general. Model-based approaches, such as the popular mixed-membership stochastic blockmodel or MMSB (Airoldi et al., 2008), use parameters for each node to specify the overlapping communities, but standard inference techniques cannot guarantee consistency. We link the two approaches, by (a) establishing sufficient conditions for the symmetric NMF optimization to have a unique solution under MMSB, and (b) proposing a computationally efficient algorithm called GeoNMF that is provably optimal and hence consistent for a broad parameter regime. We demonstrate its accuracy on both simulated and real-world datasets.

Context-Aware Recommendation Systems has gained lots of attention in both industry and academic research. Factorization Machines (FM) based recommendation has been successfully used in sparse industrial datasets for user personalized video recommendations. FM is a collaborative filtering technique for predicting a target such as user rating, given observations of interaction between some users and items. The model can incorporate any available auxiliary information about the user, the item, or the interaction which serves as context or features of the data. In this paper, we propose a framework to automatically select features on FM-based recommender systems to improve the prediction quality. FM requires the input data as a one-hot encoded feature vector in a binary space domain. We use the values of the FM parameters in the binary space to determine the importance of the context. We consider the density of the important features in the binary space to rank and select the relevant features in the original data. Experiments on multiple datasets have been conducted to validate the efficiency and robustness of our method.

With the pervasion of mobile devices, crowdsourcing based received signal strength (RSS) fingerprint collection method has drawn much attention to facilitate the indoor localization since it is effective and requires no pre-deployment. However, in large open indoor environment like museums and exhibition centres, RSS measurement points cannot be collocated densely, which degrades localization accuracy. This paper focuses on measurement point collocation in different cases and their effects on localization accuracy. We first study two simple preliminary cases under assumption that users are uniformly distributed: when measurement points are collocated regularly, we propose a collocation pattern which is most beneficial to localization accuracy; when measurement points are collocated randomly, we prove that localization accuracy is limited by a tight bound. Under the general case that users are distributed asymmetrically, we show the best allocation scheme of measurement points: measurement point density ρ is proportional to (cμ)2/3in every part of the region, where μ is user density and c is a constant determined by the collocation pattern. We also give some guidelines on collocation choice and perform extensive simulations to validate our assumptions and results.

Crowdsourcing is an effective paradigm in human centric computing for addressing problems by utilizing human computation power. While efforts have been made to study the crowdsourcing systems for labeling tasks such as classification, those for scoring tasks with continuous and correlative answers have not been well studied. In this paper, we propose two inference algorithms, MCE (Maximum Correlation Estimate) and WMCE (Weighted Maximum Correlation Estimate), to infer true answers based on answers submitted by workers. When estimating answers, WMCE algorithm assigns diverse weight to submitted answers of workers based on their quality while MCE algorithm assigns identical weight to submitted answers of all workers. For a fixed worker population, we reveal that the increase in task redundancy1can improve accuracy of estimated answers but such improvement is limited within a certain level. We further show that WMCE algorithm can reduce the influence of this limitation better than MCE algorithm for the same crowdsourcing system. Simulation results validate our theoretical analysis and show that WMCE algorithm outperforms MCE algorithm in the accuracy of estimated answers.