1 March 2015 in Information and Software Technology
27 May 2019 in IEEE Transactions on Services Computing
Next point-of-interest (POI) recommendation is recently proposed to predict user's next destination and has attracted considerable research interest. Most of the previous studies failed to incorporate the spatiotemporal contextual information, which plays a critical role in analyzing user check-in behavior, into recommending the next POI. In recent years, embedding learning and recurrent neural network (RNN) based approaches show promising performance for modeling sequential patterns of check-in behavior in next POI recommendation. However, not all of the historical check-in records contribute equally to the next-step check-in behavior. To provide better next POI recommendation performance, we first proposed a spatiotemporal long and short-term memory (ST-LSTM) network. By feeding the spatiotemporal contextual information into the LSTM network in each step, ST-LSTM can model the spatial and temporal information better. Also, we developed an attention-based spatiotemporal LSTM (ATST-LSTM) network for next POI recommendation. By using the attention mechanism, ATST-LSTM can focus on the relevant historical check-in records in a check-in sequence selectively using the spatiotemporal contextual information. Besides, we conducted a comprehensive performance evaluation using large-scale real-world datasets collected from two popular location-based social networks. Experimental results indicated that the proposed ATST-LSTM network outperformed two state-of-the-art next POI recommendation approaches regarding three evaluation metrics.
1 October 2019 in IEEE Journal of Biomedical and Health Informatics
Uterine cancer (also known as endometrial cancer) can seriously affect the female reproductive system, and histopathological image analysis is the gold standard for diagnosing endometrial cancer. Due to the limited ability to model the complicated relationships between histopathological images and their interpretations, existing computer-aided diagnosis (CADx) approaches using traditional machine learning algorithms often failed to achieve satisfying results. In this study, we develop a CADx approach based on a convolutional neural network (CNN) and attention mechanisms, called HIENet. In the ten-fold cross-validation on ~3,300 hematoxylin and eosin (H&E) image patches from ~500 endometrial specimens, HIENet achieved a 76.91 ± 1.17% (mean ± s. d.) accuracy for four classes of endometrial tissue, i.e., normal endometrium, endometrial polyp, endometrial hyperplasia, and endometrial adenocarcinoma. Also, HIENet obtained an area-under-the-curve (AUC) of 0.9579 ± 0.0103 with an 81.04 ± 3.87% sensitivity and 94.78 ± 0.87% specificity in a binary classification task that detected endometrioid adenocarcinoma. Besides, in the external validation on 200 H&E image patches from 50 randomly-selected female patients, HIENet achieved an 84.50% accuracy in the four-class classification task, as well as an AUC of 0.9829 with a 77.97% (95% confidence interval, CI, 65.27%~87.71%) sensitivity and 100% (95% CI, 97.42%~100.00%) specificity. The proposed CADx method outperformed three human experts and five CNN-based classifiers regarding overall classification performance. It was also able to provide pathologists better interpretability of diagnoses by highlighting the histopathological correlations of local pixel-level image features to morphological characteristics of endometrial tissue.
15 January 2020 in IEEE Transactions on Engineering Management
As service-oriented computing (SOC) technologies gradually mature, developing service-based systems (such as mashups) has become increasingly popular in recent years. Faced with the rapidly increasing number of Web services, recommending appropriate component services for developers on demand is a vital issue in the development of mashups. In particular, since a new mashup to develop contains no component services, it is a new “user” to a service recommender system. To address this new “user” cold-start problem, we propose a multiplex interaction-oriented service recommendation approach, named MISR, which incorporates three types of interactions between services and mashups into a deep neural network. In this article, we utilize the powerful representation learning abilities provided by deep learning to extract hidden structures and features from various types of interactions between mashups and services. Experiments conducted on a real-world dataset from ProgrammableWeb show that MISR outperforms several state-of-the-art approaches regarding commonly used evaluation metrics.
6 December 2011 in Journal of Systems Science and Complexity
6 August 2013 in Optics Express
JRJanarthanan Rasakanthan
High speed, high resolution and high sensitivity are desirable for optical coherence tomography (OCT). Here, we demonstrate a space-division multiplexing (SDM) technology that translates long coherence length of a commercially available wavelength tunable laser into high OCT imaging speed. We achieved an effective 800,000 A-scans/s imaging speed using a 100,000 Hz tunable vertical cavity surface-emitting laser (VCSEL). A sensitivity of 94.6 dB and a roll-off of < 2 dB over ~30 mm imaging depth were measured from a single channel in the prototype SDM-OCT system. An axial resolution of ~11 μm in air (or ~8.3 μm in tissue) was achieved throughout the entire depth range. An in vivo, 3D SDM-OCT volume of an entire Drosophila larva consisting of 400 x 605 A-scans was acquired in 0.37 seconds. Synchronized cross-sectional OCT imaging of three different segments of a beating Drosophila larva heart is demonstrated. The SDM technology provides a new orthogonal dimension for further speed improvement for OCT with favorable cost scaling. SDM-OCT also preserves image resolution and allows synchronized cross-sectional and three-dimensional (3D) imaging of biological samples, enabling new biomedical applications.