Fran Casino

On Privacy Preserving Collaborative Filtering: Current Trends, Open Problems, and New Issues

Automatic recommender systems have become a cornerstone of e-commerce, especially after the great welcome of Web 2.0 based on participation and interaction of Internet users. Collaborative Filtering (CF) is a recommender system that is becoming increasingly relevant for the industry due to the growth of the Internet, which has made it much more difficult to effectively extract useful information. In this paper, we introduce a taxonomy of the different CF families and we discuss the most relevant Privacy Preserving Collaborative Filtering (PPCF) methods in the literature. To understand the inherent challenges of the PPCF, we also conduct an overview of the current tendencies and major drawbacks of this kind of recommender systems, and we propose several strategies to overcome the shortcomings.

Privacy Preserving Collaborative Filtering with k-Anonymity through Microaggregation

Collaborative Filtering (CF) is a recommender system which is becoming increasingly relevant for the industry. Current research focuses on Privacy Preserving Collaborative Filtering (PPCF), whose aim is to solve the privacy issues raised by the systematic collection of private information. In this paper, we propose a new micro aggregation-based PPCF method that distorts data to provide k-anonymity, whilst simultaneously making accurate recommendations. Experimental results demonstrate that the proposed method perturbs data more efficiently than the well-known and widely used distortion method based on Gaussian noise addition.

Optimized Wireless Channel Characterization in Large Complex Environments by Hybrid Ray Launching-Collaborative Filtering Approach

Simulation techniques based on deterministic methods such as ray tracing and ray launching are widely used to perform radioplanning tasks. However, the quality of the simulations depends on the number of rays and the angular resolution. The computational cost of these simulations in high definition prevents their use in complex environment, and their low-definition counterparts are used instead. In this letter, we propose a technique based on collaborative filtering to lessen the poor-quality problems of low-definition simulations. We show that our approach obtains results very similar to those of high definition in much less time. Also, we compare our approach to other well-known techniques and show that it performs better in terms of accuracy and precision. The use of combined deterministic/collaborative filtering techniques allows the estimation of radioplanning tasks in large, complex scenarios with a potentially large amount of transceivers.

A study on the detection of wandering patterns in human trajectories

Wandering and disorientation are problems usually suffered by people with dementia. Automatically detecting these mobility patterns is a complex task that has captured the attention of the research community for a long time. Several research projects have been devoted to the analysis of those patterns and their relation with Alzheimer and other cognitive disorders. However, most of them have been focussed on indoor environments. In this paper, we address the problem of detecting mobility patterns associated with wandering and disorientation in outdoor scenarios. To do so, we have analysed the key features of wandering trajectories (specially tailored to this end) and, we have compared them with real trajectories in which wandering was not observed. We have found that, by using some key indicators, significant differences between wandering and normal trajectories can be observed, thus allowing the detection of wandering trajectories in real outdoor scenarios.

Performance Analysis of ZigBee Wireless Networks for AAL through Hybrid Ray Launching and Collaborative Filtering

This paper presents a novel hybrid simulation method based on the combination of an in-house developed 3D ray launching algorithm and a collaborative filtering (CF) technique, which will be used to analyze the performance of ZigBee-based wireless sensor networks (WSNs) to enable ambient assisted living (AAL). The combination of Low Definition results obtained by means of a deterministic ray launching method and the application of a CF technique leads to a drastic reduction of the time and computational cost required to obtain accurate simulation results. The paper also reports that this kind of AAL indoor complex scenario with multiple wireless devices needs a thorough and personalized radioplanning analysis as radiopropagation has a strong dependence on the network topology and the specific morphology of the scenario. The wireless channel analysis performed by our hybrid method provides valuable insight into network design phases of complex wireless systems, typical in AAL-oriented environments. Thus, it results in optimizing network deployment, reducing overall interference levels, and increasing the overall system performance in terms of cost reduction, transmission rates, and energy efficiency.

Context-aware recommender for smart health

Smart cities are equipped with sensors and communications infrastructures that can provide lots of data. In addition, citizens might use their smartphones to share information with the city and with other users. With the aim to use the infrastructure of smart cities to improve the quality of life of citizens and their healthcare system, the concept of smart health (s-health) was born. In this article we show how recommender systems could be used to provide healthcare services within the context of a smart city in which citizens collaborate with the city to improve their quality of life.

Dense wireless sensor network design for the implementation of Smart Health environments

In this work, the analysis on the design and implementation of dense wireless sensor networks is presented, by performing deterministic wireless channel analysis, providing assessment on topological and morphological impact of complex urban environments in the operation of the employed wireless sensor networks, within WLAN/WPAN standards. The proposed methodology, combining initial radioplanning estimations aided with complete scenario description can provide a useful tool in the deployment of future massive wireless networks, providing optimized levels of service whilst minimizing energy consumption and overall interference levels.

An Autonomous System to Assess, Display and Communicate the Pain Level in Newborns

Pain is an issue that medicine considers of great importance. The treatment of pain and discomfort is essential during hospitalisation procedures, specially for newborn infants because, first, they are not able to communicate that they are in pain; and second large periods of pain or discomfort can lead to major issues. In order to assess the pain/discomfort suffered by patients, healthcare professionals use pain assessment scales: using physiologic parameters (e.g., heart beat, blood oxygen level, etc.) and observed behavioural parameters (e.g., cry, spasmodic movements, etc.) a value for pain is obtained. Current methods for pain assessment have some drawbacks, which could be overcome with the use of computerised systems. With this aim, we present a system that automatically analyses the pain or discomfort levels of newborns. The proposed system allows the remote monitoring of newborns and raises alarms upon specific conditions. Hence, caregivers (e.g., nurses and their assistants) can act accordingly to help relieving the pain. Moreover, the remotely monitoring is also useful for parents, since they cannot stay close to their babies hospitalised in neonate Intensive Care Units.

Healthy Routes in the Smart City: A Context-Aware Mobile Recommender

Smart cities have sensors and communication infrastructures that provide humongous amounts of data. People can extend this infrastructure, acting as independent, mobile sensors through their smartphones, enabling opportunistic sensing. This information provides locality and timeliness of measurements that would otherwise be unavailable. Researchers have developed a context-aware healthy-route recommender system that offers personalized recommendations to people according to their medical condition and real-time information from the smart city. Experiments with a simulated dataset and real data verified the system’s usefulness.

Wandering detection methods in smart cities: Current and new approaches

People suffering from mild cognitive impairments and early stages of dementia may notice a deterioration in their memory functions. Consequently, disorientation or wanderinglike behaviours might occur in their daily activities. Detecting wandering in trajectories is a complex task highly influenced by the technology/ies used and the context in which people move. Hence, there is no commonly accepted technique to detect wandering automatically. Since technology is a key factor, Smart Cities can open the door to new opportunities to approach the problem. In this article, we briefly summarise the state of the art of wandering detection techniques, we describe some of the benefits that Smart Cities will contribute with, and we provide a preliminary proposal of a new wandering detection method.