scholarly journals Machine Learning and Soil Humidity Sensing: Signal Strength Approach

2022 ◽  
Vol 22 (2) ◽  
pp. 1-21
Author(s):  
Lea Dujić Rodić ◽  
Tomislav Županović ◽  
Toni Perković ◽  
Petar Šolić ◽  
Joel J. P. C. Rodrigues
Keyword(s):  
Machine Learning ◽  
Signal Strength ◽  
Cost Effective ◽  
Wireless Channel ◽  
Soil Humidity ◽  
Humidity Sensing ◽  
Learning Techniques ◽  
Effective System ◽  
Ubiquitous And Pervasive Computing ◽  
The Given

The Internet-of-Things vision of ubiquitous and pervasive computing gives rise to future smart irrigation systems comprising the physical and digital worlds. A smart irrigation ecosystem combined with Machine Learning can provide solutions that successfully solve the soil humidity sensing task in order to ensure optimal water usage. Existing solutions are based on data received from the power hungry/expensive sensors that are transmitting the sensed data over the wireless channel. Over time, the systems become difficult to maintain, especially in remote areas due to the battery replacement issues with a large number of devices. Therefore, a novel solution must provide an alternative, cost- and energy-effective device that has unique advantage over the existing solutions. This work explores the concept of a novel, low-power, LoRa-based, cost-effective system that achieves humidity sensing using Deep Learning techniques that can be employed to sense soil humidity with high accuracy simply by measuring the signal strength of the given underground beacon device.

scholarly journals Sensing Occupancy through Software: Smart Parking Proof of Concept

Electronics ◽  
2020 ◽  
Vol 9 (12) ◽  
pp. 2207
Author(s):  
Lea Dujić Rodić ◽  
Toni Perković ◽  
Tomislav Županović ◽  
Petar Šolić
Keyword(s):  
Signal Strength ◽  
Cost Effective ◽  
Receiver Side ◽  
Strength Change ◽  
Multiple Components ◽  
The Neural Network ◽  
Presence Data ◽  
Processing Signal ◽  
Smart Parking ◽  
The Given

In order to detect the vehicle presence in parking slots, different approaches have been utilized, which range from image recognition to sensing via detection nodes. The last one is usually based on getting the presence data from one or more sensors (commonly magnetic or IR-based), controlled and processed by a micro-controller that sends the data through radio interface. Consequently, given nodes have multiple components, adequate software is required for its control and state-machine to communicate its status to the receiver. This paper presents an alternative, cost-effective beacon-based mechanism for sensing the vehicle presence. It is based on the well-known effect that, once the metallic obstacle (i.e., vehicle) is on top of the sensing node, the signal strength will be attenuated, while the same shall be recognized at the receiver side. Therefore, the signal strength change conveys the information regarding the presence. Algorithms processing signal strength change at the receiver side to estimate the presence are required due to the stochastic nature of signal strength parameters. In order to prove the concept, experimental setup based on LoRa-based parking sensors was used to gather occupancy/signal strength data. In order to extract the information of presence, the Hidden Markov Model (HMM) was employed with accuracy of up to 96%, while the Neural Network (NN) approach reaches an accuracy of up to 97%. The given approach reduces the costs of the sensor production by at least 50%.

Borderline and Depression: A Thin EEG Line

2021 ◽  
pp. 155005942110608
Author(s):  
Jakša Vukojević ◽  
Damir Mulc ◽  
Ivana Kinder ◽  
Eda Jovičić ◽  
Krešimir Friganović ◽  
...  
Keyword(s):  
Machine Learning ◽  
Borderline Personality ◽  
Machine Learning Techniques ◽  
Major Depressive ◽  
Everyday Clinical Practice ◽  
Data Set ◽  
Learning Techniques ◽  
Eeg Recordings ◽  
The Given ◽  
Close Interrelationship

In everyday clinical practice, there is an ongoing debate about the nature of major depressive disorder (MDD) in patients with borderline personality disorder (BPD). The underlying research does not give us a clear distinction between those 2 entities, although depression is among the most frequent comorbid diagnosis in borderline personality patients. The notion that depression can be a distinct disorder but also a symptom in other psychopathologies led our team to try and delineate those 2 entities using 146 EEG recordings and machine learning. The utilized algorithms, developed solely for this purpose, could not differentiate those 2 entities, meaning that patients suffering from MDD did not have significantly different EEG in terms of patients diagnosed with MDD and BPD respecting the given data and methods used. By increasing the data set and the spatiotemporal specificity, one could have a more sensitive diagnostic approach when using EEG recordings. To our knowledge, this is the first study that used EEG recordings and advanced machine learning techniques and further confirmed the close interrelationship between those 2 entities.

An Insight of Machine Learning in Web Network Analysis

2019 ◽  
Vol 11 (2) ◽  
pp. 20-34
Author(s):  
Meenakshi Sharma ◽  
Anshul Garg
Keyword(s):  
Machine Learning ◽  
Machine Learning Techniques ◽  
Effective Solution ◽  
Business Organisation ◽  
Web Structure ◽  
Learning Techniques ◽  
Effective Decision ◽  
Effective Decision Making ◽  
The Given ◽  
The Web

The World Wide Web is immensely rich in knowledge. The knowledge comes from both the content and distinctive characteristics of the web like its hyperlink structure. The problem comes in digging the relevant data from the web and giving the most appropriate decision to solve the given problem, which can be used for improving any business organisation. The effective solution of the problem depends on how efficiently and effectively the analysis of the web data is done. In analysing the data on web, not only relevant content analysis is essential but also the analysis of web structure is important. This article gives a brief introduction about the various terminologies and measures like centrality, Page Rank, and density used in the web networking analysis. This article will also give a brief introduction about the various supervised ML techniques such as classification, regression, and unsupervised machine learning techniques such as clustering, etc., which are very useful in analysing the web network so that user can make quick and effective decision making

scholarly journals Developing an Open-Source Lightweight Game Engine with DNN Support

Electronics ◽  
2020 ◽  
Vol 9 (9) ◽  
pp. 1421
Author(s):  
Haechan Park ◽  
Nakhoon Baek
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Open Source ◽  
Programming Languages ◽  
Cost Effective ◽  
Machine Learning Techniques ◽  
Learning Technology ◽  
Game Engine ◽  
Learning Techniques ◽  
Technical Issues

With the growth of artificial intelligence and deep learning technology, we have many active research works to apply the related techniques in various fields. To test and apply the latest machine learning techniques in gaming, it will be very useful to have a light-weight game engine for quick prototyping. Our game engine is implemented in a cost-effective way, in comparison to well-known commercial proprietary game engines, by utilizing open source products. Due to its simple internal architecture, our game engine is especially beneficial for modifying and reviewing the new functions through quick and repetitive tests. In addition, the game engine has a DNN (deep neural network) module, with which the proposed game engine can apply deep learning techniques to the game features, through applying deep learning algorithms in real-time. Our DNN module uses a simple C++ function interface, rather than additional programming languages and/or scripts. This simplicity enables us to apply machine learning techniques more efficiently and casually to the game applications. We also found some technical issues during our development with open sources. These issues mostly occurred while integrating various open source products into a single game engine. We present details of these technical issues and our solutions.

From information- to knowledge-management: the role of rule induction and neural net machine learning techniques in knowledge generation

1989 ◽  
Vol 15 (4-5) ◽  
pp. 299-304 ◽  
Author(s):  
Nigel Ford
Keyword(s):  
Machine Learning ◽  
Cost Effective ◽  
Rule Induction ◽  
Machine Learning Techniques ◽  
Knowledge Generation ◽  
Neural Net ◽  
Knowledge Resources ◽  
Learning Techniques ◽  
Knowledge Manager

Developments in artificial intelligence mean that it is now increasingly possible to store not only information but also knowledge as an exploitable resource. Insofar as he or she is concerned with creating, organizing and monitoring knowledge resources to support effective decision making within an organization, the information manager is developing the role of knowledge manager. As well as its organization and dissemina tion, the generation of storable knowledge is very much on the agenda of the knowledge manager. The extent to which com puters can help in the process of knowledge generation is central to his or her concerns. Machine learning techniques have been developed which are capable of giving us an increasing amount of help in this process. The contributions of rule induction and artificial neural net systems are discussed. It is likely that such tech niques will prove to be useful tools both for the information/knowledge manager requiring practical working systems enabling the cost-effective exploitation of knowledge resources, and for the information/knowledge scientist requir ing advances in our more fundamental theoretical knowledge of the nature of information and ways of processing it.

scholarly journals Quantification of toxic metallic elements using machine learning techniques and spark emission spectroscopy

2019 ◽  
Author(s):  
Seyyed Ali Davari ◽  
Anthony S. Wexler
Keyword(s):  
Machine Learning ◽  
Real Time ◽  
Emission Spectroscopy ◽  
Cost Effective ◽  
The United States ◽  
Machine Learning Techniques ◽  
Metallic Elements ◽  
Learning Techniques ◽  
Single Feature ◽  
Us Epa

Abstract. The United States Environmental Protection Agency (US EPA) list of Hazardous Air Pollutants (HAPs) includes metal elements suspected or associated with development of cancer. Traditional techniques for detecting and quantifying toxic metallic elements in the atmosphere are either not real time, hindering identification of sources, or limited by instrument costs. Spark emission spectroscopy is a promising and cost effective technique that can be used for analyzing toxic metallic elements in real time. Here, we have developed a cost-effective spark emission spectroscopy system to quantify the concentration of toxic metallic elements targeted by US EPA. Specifically, Cr, Cu, Ni, and Pb solutions were diluted and deposited on the ground electrode of the spark emission system. Least Absolute Shrinkage and Selection Operator (LASSO) was optimized and employed to detect useful features from the spark-generated plasma emissions. The optimized model was able to detect atomic emission lines along with other features to build a regression model that predicts the concentration of toxic metallic elements from the observed spectra. The limits of detections (LOD) were estimated using the detected features and compared to the traditional single-feature approach. LASSO is capable of detecting highly sensitive features in the input spectrum; however for some elements the single-feature LOD marginally outperforms LASSO LOD. The combination of low cost instruments with advanced machine learning techniques for data analysis could pave the path forward for data driven solutions to costly measurements.

scholarly journals Quantification of toxic metals using machine learning techniques and spark emission spectroscopy

2020 ◽  
Vol 13 (10) ◽  
pp. 5369-5377
Author(s):  
Seyyed Ali Davari ◽  
Anthony S. Wexler
Keyword(s):  
Machine Learning ◽  
Real Time ◽  
Toxic Metals ◽  
Emission Spectroscopy ◽  
Cost Effective ◽  
The United States ◽  
Machine Learning Techniques ◽  
Learning Techniques ◽  
Single Feature ◽  
Us Epa

Abstract. The United States Environmental Protection Agency (US EPA) list of hazardous air pollutants (HAPs) includes toxic metal suspected or associated with development of cancer. Traditional techniques for detecting and quantifying toxic metals in the atmosphere are either not real time, hindering identification of sources, or limited by instrument costs. Spark emission spectroscopy is a promising and cost-effective technique that can be used for analyzing toxic metals in real time. Here, we have developed a cost-effective spark emission spectroscopy system to quantify the concentration of toxic metals targeted by the US EPA. Specifically, Cr, Cu, Ni, and Pb solutions were diluted and deposited on the ground electrode of the spark emission system. The least absolute shrinkage and selection operator (LASSO) was optimized and employed to detect useful features from the spark-generated plasma emissions. The optimized model was able to detect atomic emission lines along with other features to build a regression model that predicts the concentration of toxic metals from the observed spectra. The limits of detections (LODs) were estimated using the detected features and compared to the traditional single-feature approach. LASSO is capable of detecting highly sensitive features in the input spectrum; however, for some toxic metals the single-feature LOD marginally outperforms LASSO LOD. The combination of low-cost instruments with advanced machine learning techniques for data analysis could pave the path forward for data-driven solutions to costly measurements.

Sign in / Sign up

Export Citation Format

Share Document