Outlier detection is an integral part of data mining and has attracted much attention recently [M. Breunig et al., (2000)], [W. Jin et al., (2001)], [E. Knorr et al., (2000)]. We propose a new method for evaluating outlierness, which we call the local correlation integral (LOCI). As with the best previous methods, LOCI is highly effective for detecting outliers and groups of outliers (a.k.a. micro-clusters). In addition, it offers the following advantages and novelties: (a) It provides an automatic, data-dictated cutoff to determine whether a point is an outlier-in contrast, previous methods force users to pick cut-offs, without any hints as to what cut-off value is best for a given dataset. (b) It can provide a LOCI plot for each point; this plot summarizes a wealth of information about the data in the vicinity of the point, determining clusters, micro-clusters, their diameters and their inter-cluster distances. None of the existing outlier-detection methods can match this feature, because they output only a single number for each point: its outlierness score, (c) Our LOCI method can be computed as quickly as the best previous methods, (d) Moreover, LOCI leads to a practically linear approximate method, aLOCI (for approximate LOCI), which provides fast highly-accurate outlier detection. To the best of our knowledge, this is the first work to use approximate computations to speed up outlier detection. Experiments on synthetic and real world data sets show that LOCI and aLOCI can automatically detect outliers and micro-clusters, without user-required cut-offs, and that they quickly spot both expected and unexpected outliers.

Automated sleep stage scoring for mice is in high demand for sleep research, since manual scoring requires considerable human expertise and efforts. The existing automated scoring methods do not provide the scoring accuracy required for practical use. In addition, the performance of such methods has generally been evaluated using rather small-scale datasets, and their robustness against individual differences and noise has not been adequately verified. This research proposes a novel automated scoring method named “MC-SleepNet”, which combines two types of deep neural networks. Then, we evaluate its performance using a large-scale dataset that contains 4,200 biological signal records of mice. The experimental results show that MC-SleepNet can automatically score sleep stages with an accuracy of 96.6% and kappa statistic of 0.94. In addition, we confirm that the scoring accuracy does not significantly decrease even if the target biological signals are noisy. These results suggest that MC-SleepNet is very robust against individual differences and noise. To the best of our knowledge, evaluations using such a large-scale dataset (containing 4,200 records) and high scoring accuracy (96.6%) have not been reported in previous related studies.

Pattern matching over event streams is well developed. However, with the increasing demand of measurement accuracy, confidence of more complex events sourced from original, continuously arriving events generated from sensor kind electronic devices is becoming more and more been concerned. Actually, some applications such as RFID-based supply chain management and monitoring in health care require data stream with high reliability, but current hardware and wireless communication techniques cannot support 100% confident data, one stream processing engine which can report confidence for processed complex events over uncertain data is needed. In this paper, we propose an optimized method to not only calculate the probability of outputs of compound events but also obtain the value of confidence of the complex pattern given by user against uncertain raw input data stream generated by distrustful network devices. Our proposal is based on an existing stream processing engine SASE+, and we extend its evaluation model NFAb automaton to a new type of automaton in order to manage the runtime against probabilistic stream. In the design of automaton, we consider optimizations to reduce the computation cost and response time to a realistic degree with long sliding time window.

The location profiles of social media users are valuable for various applications, such as marketing and real-world analysis. As most users do not disclose their home locations, the problem of inferring home locations has been well studied in recent years. In fact, most existing methods perform batch inference using static (i.e., pre-stored) social media contents. However, social media contents are generated and delivered in real-time as social streams. In this situation, it is important to continuously update current inference results based on the newly arriving contents to improve the results over time. Moreover, it is effective for location inference to use the spatiotemporal correlation between contents and locations. The main idea of this paper is that we can infer the locations of users who simultaneously post about a local event (e.g., earthquakes). Hence, in this paper, we propose an online location inference method over social streams that exploits the spatiotemporal correlation, achieving 1) continuous updates with low computational and storage costs, and 2) better inference accuracy than that of existing methods. The experimental results using a Twitter dataset show that our method reduces the inference error to less than 68% of existing methods. The results also show that the proposed method can update inference results in constant time regardless of the amount of accumulated contents.