Renata Lopes Rosa

The impact of video-quality-level switching on user quality of experience in dynamic adaptive streaming over HTTP

Dynamic adaptive streaming over HTTP (DASH) has become a promising solution for video delivery services over the Internet in the last few years. Currently, several video content providers use the DASH solution to improve the users’ quality of experience (QoE) by automatically switching video quality levels (VQLs) according to the network status. However, the frequency of switching events between different VQLs during a video streaming session may disturb the user’s visual attention and therefore affect the user’s QoE. As one of the first attempts to characterize the impact of VQL switching on the user’s QoE, we carried out a series of subjective tests, which show that there is a correlation between the user QoE and the frequency, type, and temporal location of the switching events. We propose a novel parameter named switching degradation factor (SDF) to capture such correlation. A DASH algorithm with SDF parameter is compared with the same algorithm without SDF. The results demonstrate that the SDF parameter significantly improves the user’s QoE, especially when network conditions vary frequently.

Age Groups Classification in Social Network Using Deep Learning

Social networks have a large amount of data available, but often, people do not provide some of their personal data, such as age, gender, and other demographics. Although the sentiment analysis uses such data to develop useful applications in people’s daily lives, there are still failures in this type of analysis, either by the restricted number of words contained in the word dictionaries or because they do not consider the most diverse parameters that can influence the sentiments in a sentence; thus, more reliable results can be obtained, if the users profile information and their writing characteristics are considered. This research suggests that one of the most relevant parameter contained in the user profile is the age group, showing that there are typical behaviors among users of the same age group, specifically, when these users write about the same topic. A detailed analysis with 7000 sentences was performed to determine which characteristics are relevant, such as, the use of punctuation, number of characters, media sharing, topics, among others; and which ones can be disregarded for the age groups classification. Different learning machine algorithms are tested for the classification of the teenager and adult age group, and the deep convolutional neural network had the best performance, reaching a precision of 0.95 in the validation tests. Furthermore, in order to validate the usefulness of the proposed model for classifying age groups, it is implemented into the enhanced sentiment metric (eSM). In the performance validation, subjective tests are performed and the eSM with the proposed model reached a root mean square error and a Pearson correlation coefficient of 0.25 and 0.94, respectively, outperforming the eSM metric, when the age group information is not available.

Recommendation system using sentiment analysis considering the polarity of the adverb

This paper presents a recommendation system based on sentiment analysis of sentences extracted from Social Networks considering an algorithm which depends on the adverb found in the sentence. Furthermore, the recommendation system has a low complexity, therefore the perceived impacts on the analysis of energy consumption on electronic devices is almost negligible.

A Simple Method to Measure the Image Complexity on a Fault Tolerant Cluster Computing

The cluster computing is widely used for image processing in entertainment applications. Measuring the required number of nodes to be used in a cluster computing helps saving nodes processing time, even if this occurs on a fault tolerant scenario. The paper analyzes the images rendering process by ray tracing on the cluster computing using the freeware software fault tolerant message-passing interface with Povray software in a Linux-based operating system. This paper uses a simple method of image complexity measure, it is shown that depending on the images complexity level, a minimum number of cluster nodes is chose to render the image instead of using all the cluster nodes. It is useful to save machine processing time and meanwhile another image can be rendered in a parallel process with a previous one and where the solution is applied in a Fault Tolerant scenario, so independent of the network fault, the cluster continues working.

Monitoring system for potential users with depression using sentiment analysis

This paper presents a monitoring solution for users with potential psychological disturbs, specifically stress. The phrases extracted from Social Networks are filtered by mood and scored using a sentiment-analysis-metric that considers personal characteristics, such as gender and age. The solution informs, to authorized persons, the emotional changes of the users.

Voice quality assessment in mobile devices considering different fading models

This paper presents a mechanism to provide a real-time assessment of voice quality on mobile handled devices. This solution considers the voice degradations caused by the radio frequency channel and packet losses on the IP network. Furthermore, the network architecture of the proposed solution is introduced and the application is implemented on the user mobile device.

Real-time evaluation of speech quality in mobile communication services

This paper presents a mechanism to provide a real-time evaluation of speech quality. This solution consists of a voice-quality-server (VQS) based on the ITU-T P.563 algorithm with modifications to treat natural silences of human speech and packet losses on the network. Furthermore, an application is implemented on the user device.

Improving a Video Quality Metric with the Video Content Type parameter

Nowadays, there are several multimedia services, which are carried via IP networks. From these services; the traffic regarding video applications had the greatest growth in the last years. In this context, the study of methods for assessing video quality plays a very important role. In general, video quality metrics are defined from mathematical models, which use as inputs, different network or application layer parameters, and the video quality assessment results from subjective tests. This paper aims to demonstrate that the video quality metric performance can be improved by adding a parameter regarding the video content type. In order to perform video quality assessment tests in a short time period, the so-called crowdsourcing methodology is used. In this methodology remote users assess the video quality. The experimental results show that the Video streaming Quality Metric (VsQM) performance is improved when the video content type is considered (eVsQM). Indeed, the Pearson correlation factors between eVsQM and crowdsourcing results reach 0.94 ¿ 0.98, depending on the video degradation scenario.

Recommendation system based on user profile extracted from an IMS network with emphasis on social network and digital TV

This paper introduces new network architecture capable of providing a flexible and configurable Recommendation System in which network operator, subscribers and content providers could interact to each other by message exchanging. This solution considers multiple inputs provided by the user and context-based information of IP Multimedia Subsystem (IMS) network. It creates a specific subscriber profile based on information from cellular and fixed networks, digital TV and, internet services as social networks or e-mail accounts. Also, for collecting information is used a Cell Broadcast platform which can send messages to a specific geographic location achieving users with high probability of similar preferences. User feedback is collected to evaluate the utility of recommendation messages. As additional features, the Recommendation System uses presence and location information as a decision factor to send or not the recommendation messages.

New Cache System-Based Power-Aware Algorithm in MANET

The increasing use of mobile devices in wireless environments led to many studies about power-aware algorithms and protocols to improve the lifetime of device’s battery. This paper studies the implementation and testing of a new power-aware algorithm, PSWA, associated with a Cache System, and tested with the WRP routing protocol, in a Mobile Ad hoc Network scenario. A scheme that does not use the sleep schedule for the nodes it was used, nodes automatically fall asleep if they do not deal with data and routes. In the simulation scenario the WRP was chosen because it has a better performance in real-time traffic, and this type of traffic is commonly used today. The simulation results show an important decrease in energy consumption close to 10% with the association of the new power-aware algorithm.