Summary: This technical report presents a solution for monitoring water quality in river basins based on the Internet of Things, part of the project developed during the MO629 - Internet of Things - discipline at the Unicamp Institute of Computing.
Water is one of the natural resources that are essential for the survival of humanity and is included as the sixth objective in the Sustainable Development Goals (SDGs) determined by the United Nations (UN), which consists of ensuring the availability and sustainable management of drinking water and sanitation for all. However, in the context of river basins, the scarcity and finiteness of this resource is noted, due to natural issues such as low rainfall frequency, inadequate and excessive use or the return of this resource to the environment without adequate treatment. In order to achieve objective 6 of the UN SDGs, it is necessary to monitor water quality, and with the help of Information and Communication Technologies (ICT), it is possible to collect data in real time and process it so that it becomes relevant information to determine the state of water quality and then apply the necessary measures. Therefore, a solution for monitoring water quality in river basins based on the Internet of Things is presented, with the main objective of generating inputs for river basin managers so that they can take appropriate measures related to water quality and thus assisting in the guarantee established by SDG 6.
Summary: The technical report presents the Symbiotic Garden project, a ubiquitous computing initiative based on Internet of Things (IoT) technologies, developed as part of the MO629 - Internet of Things course at the Unicamp Institute of Computing. The main objective is to create an interactive environment that promotes care for local flora, in line with Sustainable Development Goal (SDG) 11, which aims at sustainable cities and communities. The system integrates sensors and actuators, such as humidity, temperature and noise meters, as well as interactive devices, such as virtual watering cans and LED screens. Communication between devices is enabled by the MQTT protocol, with local analysis of the data before it is sent to the cloud. The proposal explores the integration of technology into everyday life through human interactions, awakening sustainable behaviors and increasing environmental awareness. The project also suggests scalability to other contexts, such as community gardens, squares and parks, consolidating its application as an initial solution for the development of smart cities.
Abstract: This text presents a literature review of Neural Language Models, which are deep neural networks to encode a given language. The scope of this review covers two main topics: (i) Transformers-based Neural Networks, established as state-of-the-art in addressing Natural Language Processing (NLP) problems and a suitable approach to train Language Models; and (ii) Neural Language Models that compress the statistical semantics of textual data into word vectors. These word vectors computationally represent the basic units of the language at hand. In fact, obtaining a computational representation for textual constructs is a long-standing problem that has challenged diverse NLP approaches. We analyzed the usage of language models for Topic Modeling and for Semantic Annotation of Virtual Patients. The establishment of transformers-based language models opens up vast possibilities and perspectives on interdisciplinary topics. This text concludes with a critical analysis addressing issues regarding applications based on language models.
Instituto de Computação :: State University of Campinas
Av. Albert Einstein, 1251 - Cidade Universitária Zeferino Vaz • 13083-852 Campinas, SP - Brazil • Phone: [19] 3521-5838 |