Dear Colleagues,

Pervasive electronics provide an enabling technology supporting intelligent processing systems. The availability of distributed sensing and smart devices is changing the way we approach problems. Sentiment analysis is a landmark example due to its inherent complexity. Researchers traditionally approached this problem using text data; now, we have text, images, videos, audio information, biomedical data, information about the user's position, and data from smart devices. This information can boost sentiment analysis tools, opening up new scenarios.

The effective performance of sentiment analysis depends on two components: a sensing system that can provide accurate information about the user's state of mind, and an intelligent processing system that can utilize such information. The latter aspect can be addressed by machine learning (ML), which can mine information from data at the expense of intensive computation. In particular, deep learning automates the feature-extraction process. Then, new sources of information provided by sensors can be exploited with limited human intervention. However, the amount of data is such that server-based solutions are not enough. Distributed computing solutions overcome the limitation imposed by the server-based computing paradigm by splitting the computational load into nodes that range from servers to embedded systems. The development of this eld necessarily passes through the use of software–hardware co-design.

Using embedded systems to process new data sources is indeed becoming a requirement for building the next generation of sentiment analysis systems. On the other hand, given the constraints imposed by embedded devices in terms of power consumption, latency, size, and cost, the deployment of an ML model on an embedded system poses major challenges. The main goal is to prompt frame-efficient inference functions that can run on resource-constrained edge devices. Under such a paradigm, training might, in principle, be demanded for a different, more powerful platforms. A more demanding goal is to be able to complete training on resource-constrained devices.

This Special Issue will focus on software and hardware models and methodologies for sentiment analysis. The aim is to collect the most recent advances in machine learning research for sentiment analysis. Accordingly, the Special Issue welcomes methods and ideas that emphasize the impact of embedded machine learning and novel sensor sources on sentiment analysis technologies and the use of new sensing methods to detect the human state of mind.

The topics of interest for this Special Issue include, but are not limited to:

• Software/hardware techniques for sentiment analysis;
• Sentiment analysis using IoT data;
• Embedded machine learning;
• Low-power inference engines;
• Intelligent sensors;
• Online learning on resource-constrained edge devices;
• Power-efficient machine learning implementations on embedded devices;
• The on-chip training of deep neural networks;
• High-performance, low-power computing for deep learning and computer vision;
• High-performance, low-power computing for deep-learning-based audio and speech processing;
• Machine learning for sentiment-aware autonomous systems.

Dr. Edoardo Ragusa
Prof. Erik Cambria
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• Software/hardware techniques for sentiment analysis
• Sentiment analysis using IoT data
• Embedded machine learning
• Low-power inference engines
• Intelligent sensors
• Online learning on resource-constrained edge devices
• Power-efficient machine learning implementations on embedded devices
• The on-chip training of deep neural networks
• High-performance, low-power computing for deep learning and computer vision
• High-performance, low-power computing for deep-learning-based audio and speech processing
• Machine learning for sentiment-aware autonomous systems