Two-Phase Sensor Decision: Machine-Learning for Bird Sound Recognition and Vineyard Protection

In the digital microfluidic experiments, the droplet characteristics and flow patterns are generally identified and predicted by the empirical methods, which are difficult to process a large amount of data mining. In addition, due to the existence of inevitable human invention, the inconsistent judgment standards make the comparison between different experiments cumbersome and almost impossible. In this paper, we tried to use machine learning to build algorithms that could automatically identify, judge, and predict flow patterns and droplet characteristics, so that the empirical judgment was transferred to be an intelligent process. The difference on the usual machine learning algorithms, a generalized variable system was introduced to describe the different geometry configurations of the digital microfluidics. Specifically, Buckingham’s theorem had been adopted to obtain multiple groups of dimensionless numbers as the input variables of machine learning algorithms. Through the verification of the algorithms, the SVM and BPNN algorithms had classified and predicted the different flow patterns and droplet characteristics (the length and frequency) successfully. By comparing with the primitive parameters system, the dimensionless numbers system was superior in the predictive capability. The traditional dimensionless numbers selected for the machine learning algorithms should have physical meanings strongly rather than mathematical meanings. The machine learning algorithms applying the dimensionless numbers had declined the dimensionality of the system and the amount of computation and not lose the information of primitive parameters.

Download Full-text

Thermodynamics and Machine Learning Based Approaches for Vapor–Liquid–Liquid Phase Equilibria in n-Octane/Water, as a Naphtha–Water Surrogate in Water Blends

Processes ◽

10.3390/pr9030413 ◽

2021 ◽

Vol 9 (3) ◽

pp. 413

Author(s):

Sandra Lopez-Zamora ◽

Jeonghoon Kong ◽

Salvador Escobedo ◽

Hugo de Lasa

Keyword(s):

Machine Learning ◽

Liquid Phase ◽

Phase Equilibria ◽

Aspen Plus ◽

Phase Region ◽

Two Phase ◽

Technical Literature ◽

Phase Liquid ◽

Liquid Vapor ◽

Vapor Liquid

The prediction of phase equilibria for hydrocarbon/water blends in separators, is a subject of considerable importance for chemical processes. Despite its relevance, there are still pending questions. Among them, is the prediction of the correct number of phases. While a stability analysis using the Gibbs Free Energy of mixing and the NRTL model, provide a good understanding with calculation issues, when using HYSYS V9 and Aspen Plus V9 software, this shows that significant phase equilibrium uncertainties still exist. To clarify these matters, n-octane and water blends, are good surrogates of naphtha/water mixtures. Runs were developed in a CREC vapor–liquid (VL_ Cell operated with octane–water mixtures under dynamic conditions and used to establish the two-phase (liquid–vapor) and three phase (liquid–liquid–vapor) domains. Results obtained demonstrate that the two phase region (full solubility in the liquid phase) of n-octane in water at 100 °C is in the 10-4 mol fraction range, and it is larger than the 10-5 mol fraction predicted by Aspen Plus and the 10-7 mol fraction reported in the technical literature. Furthermore, and to provide an effective and accurate method for predicting the number of phases, a machine learning (ML) technique was implemented and successfully demonstrated, in the present study.

Download Full-text

Machine learning based models for pressure drop estimation of two-phase adiabatic air-water flow in micro-finned tubes: Determination of the most promising dimensionless feature set

Chemical Engineering Research and Design ◽

10.1016/j.cherd.2021.01.002 ◽

2021 ◽

Vol 167 ◽

pp. 252-267

Author(s):

Behzad Najafi ◽

Keivan Ardam ◽

Andrej Hanušovský ◽

Fabio Rinaldi ◽

Luigi Pietro Maria Colombo

Keyword(s):

Machine Learning ◽

Pressure Drop ◽

Water Flow ◽

Two Phase ◽

Finned Tubes

Download Full-text

Distributed numerical and machine learning computations via two-phase execution of aggregated join trees

Proceedings of the VLDB Endowment ◽

10.14778/3450980.3450991 ◽

2021 ◽

Vol 14 (7) ◽

pp. 1228-1240

Author(s):

Dimitrije Jankov ◽

Binhang Yuan ◽

Shangyu Luo ◽

Chris Jermaine

Keyword(s):

Machine Learning ◽

Query Execution ◽

Relational System ◽

Numerical Computations ◽

Two Phase ◽

Execution Strategy ◽

Relational Joins

When numerical and machine learning (ML) computations are expressed relationally, classical query execution strategies (hash-based joins and aggregations) can do a poor job distributing the computation. In this paper, we propose a two-phase execution strategy for numerical computations that are expressed relationally, as aggregated join trees (that is, expressed as a series of relational joins followed by an aggregation). In a pilot run, lineage information is collected; this lineage is used to optimally plan the computation at the level of individual records. Then, the computation is actually executed. We show experimentally that a relational system making use of this two-phase strategy can be an excellent platform for distributed ML computations.

Download Full-text

Implementation of Machine Learning Algorithms for Prediction of Fluidelastic Instability in Tube Arrays

Journal of Pressure Vessel Technology ◽

10.1115/1.4049876 ◽

2021 ◽

Vol 143 (2) ◽

Author(s):

Joaquin E. Moran ◽

Yasser Selima

Keyword(s):

Machine Learning ◽

Logistic Regression ◽

Learning Algorithm ◽

Machine Learning Algorithms ◽

Two Phase ◽

Factors Affecting ◽

Logistic Regression Models ◽

Number Of Factors ◽

Tube Arrays ◽

Fluidelastic Instability

Abstract Fluidelastic instability (FEI) in tube arrays has been studied extensively experimentally and theoretically for the last 50 years, due to its potential to cause significant damage in short periods. Incidents similar to those observed at San Onofre Nuclear Generating Station indicate that the problem is not yet fully understood, probably due to the large number of factors affecting the phenomenon. In this study, a new approach for the analysis and interpretation of FEI data using machine learning (ML) algorithms is explored. FEI data for both single and two-phase flows have been collected from the literature and utilized for training a machine learning algorithm in order to either provide estimates of the reduced velocity (single and two-phase) or indicate if the bundle is stable or unstable under certain conditions (two-phase). The analysis included the use of logistic regression as a classification algorithm for two-phase flow problems to determine if specific conditions produce a stable or unstable response. The results of this study provide some insight into the capability and potential of logistic regression models to analyze FEI if appropriate quantities of experimental data are available.

Download Full-text

Detecting Malicious Traffic through Two-phase Machine Learning

Proceedings of the Asia-Pacific Advanced Network ◽

10.7125/40.6 ◽

2015 ◽

Vol 40 (0) ◽

pp. 34

Author(s):

Kazuma Shinomiya ◽

Shigeki Goto

Keyword(s):

Machine Learning ◽

Two Phase

Download Full-text

A Deep Machine Learning Method for Concurrent and Interleaved Human Activity Recognition

Sensors ◽

10.3390/s20205770 ◽

2020 ◽

Vol 20 (20) ◽

pp. 5770 ◽

Cited By ~ 1

Author(s):

Keshav Thapa ◽

Zubaer Md. Abdullah Al ◽

Barsha Lamichhane ◽

Sung-Hyun Yang

Keyword(s):

Machine Learning ◽

Random Field ◽

Activity Recognition ◽

Human Activity ◽

Short Term Memory ◽

Conditional Random Field ◽

Human Activity Recognition ◽

Second Phase ◽

Complex Activity ◽

Two Phase

Human activity recognition has become an important research topic within the field of pervasive computing, ambient assistive living (AAL), robotics, health-care monitoring, and many more. Techniques for recognizing simple and single activities are typical for now, but recognizing complex activities such as concurrent and interleaving activity is still a major challenging issue. In this paper, we propose a two-phase hybrid deep machine learning approach using bi-directional Long-Short Term Memory (BiLSTM) and Skip-Chain Conditional random field (SCCRF) to recognize the complex activity. BiLSTM is a sequential generative deep learning inherited from Recurrent Neural Network (RNN). SCCRFs is a distinctive feature of conditional random field (CRF) that can represent long term dependencies. In the first phase of the proposed approach, we recognized the concurrent activities using the BiLSTM technique, and in the second phase, SCCRF identifies the interleaved activity. Accuracy of the proposed framework against the counterpart state-of-art methods using the publicly available datasets in a smart home environment is analyzed. Our experiment’s result surpasses the previously proposed approaches with an average accuracy of more than 93%.

Download Full-text

Environmental Sound Recognition on Embedded Systems: From FPGAs to TPUs

Electronics ◽

10.3390/electronics10212622 ◽

2021 ◽

Vol 10 (21) ◽

pp. 2622

Author(s):

Jurgen Vandendriessche ◽

Nick Wouters ◽

Bruno da Silva ◽

Mimoun Lamrini ◽

Mohamed Yassin Chkouri ◽

...

Keyword(s):

Machine Learning ◽

High Performance ◽

Machine Learning Techniques ◽

Sound Recognition ◽

Learning Approaches ◽

Environmental Sound ◽

Embedded Devices ◽

Power Efficient ◽

Computationally Intensive ◽

Environmental Sound Recognition

In recent years, Environmental Sound Recognition (ESR) has become a relevant capability for urban monitoring applications. The techniques for automated sound recognition often rely on machine learning approaches, which have increased in complexity in order to achieve higher accuracy. Nonetheless, such machine learning techniques often have to be deployed on resource and power-constrained embedded devices, which has become a challenge with the adoption of deep learning approaches based on Convolutional Neural Networks (CNNs). Field-Programmable Gate Arrays (FPGAs) are power efficient and highly suitable for computationally intensive algorithms like CNNs. By fully exploiting their parallel nature, they have the potential to accelerate the inference time as compared to other embedded devices. Similarly, dedicated architectures to accelerate Artificial Intelligence (AI) such as Tensor Processing Units (TPUs) promise to deliver high accuracy while achieving high performance. In this work, we evaluate existing tool flows to deploy CNN models on FPGAs as well as on TPU platforms. We propose and adjust several CNN-based sound classifiers to be embedded on such hardware accelerators. The results demonstrate the maturity of the existing tools and how FPGAs can be exploited to outperform TPUs.

Download Full-text

Intra-domain and cross-domain transfer learning for time series

10.5194/egusphere-egu21-12142 ◽

2021 ◽

Author(s):

Erik Otović ◽

Marko Njirjak ◽

Dario Jozinović ◽

Goran Mauša ◽

Alberto Michelini ◽

...

Keyword(s):

Machine Learning ◽

Time Series ◽

Transfer Learning ◽

Time Series Data ◽

The Other ◽

Series Data ◽

Sound Recognition ◽

Transfer Of Knowledge ◽

Learning Models ◽

Machine Learning Models

In this study, we compared the performance of machine learning models trained using transfer learning and those that were trained from scratch - on time series data. Four machine learning models were used for the experiment. Two models were taken from the field of seismology, and the other two are general-purpose models for working with time series data. The accuracy of selected models was systematically observed and analyzed when switching within the same domain of application (seismology), as well as between mutually different domains of application (seismology, speech, medicine, finance). In seismology, we used two databases of local earthquakes (one in counts, and the other with the instrument response removed) and a database of global earthquakes for predicting earthquake magnitude; other datasets targeted classifying spoken words (speech), predicting stock prices (finance) and classifying muscle movement from EMG signals (medicine). In practice, it is very demanding and sometimes impossible to collect datasets of tagged data large enough to successfully train a machine learning model. Therefore, in our experiment, we use reduced data sets of 1,500 and 9,000 data instances to mimic such conditions. Using the same scaled-down datasets, we trained two sets of machine learning models: those that used transfer learning for training and those that were trained from scratch. We compared the performances between pairs of models in order to draw conclusions about the utility of transfer learning. In order to confirm the validity of the obtained results, we repeated the experiments several times and applied statistical tests to confirm the significance of the results. The study shows when, within the set experimental framework, the transfer of knowledge brought improvements in terms of model accuracy and in terms of model convergence rate. Our results show that it is possible to achieve better performance and faster convergence by transferring knowledge from the domain of global earthquakes to the domain of local earthquakes; sometimes also vice versa. However, improvements in seismology can sometimes also be achieved by transferring knowledge from medical and audio domains. The results show that the transfer of knowledge between other domains brought even more significant improvements, compared to those within the field of seismology. For example, it has been shown that models in the field of sound recognition have achieved much better performance compared to classical models and that the domain of sound recognition is very compatible with knowledge from other domains. We came to similar conclusions for the domains of medicine and finance. Ultimately, the paper offers suggestions when transfer learning is useful, and the explanations offered can provide a good starting point for knowledge transfer using time series data.

Download Full-text

Machine Learning Enhancing Adaptivity of Multimodal Mobile Systems

Machine Learning ◽

10.4018/978-1-60960-818-7.ch414 ◽

2012 ◽

pp. 969-985

Author(s):

Floriana Esposito ◽

Teresa M.A. Basile ◽

Nicola Di Mauro ◽

Stefano Ferilli

Keyword(s):

Machine Learning ◽

Mobile Device ◽

Pattern Discovery ◽

User Interaction ◽

User Model ◽

Mobile Systems ◽

Machine Learning Techniques ◽

Two Phase ◽

Learning Techniques ◽

First Time

One of the most important features of a mobile device concerns its flexibility and capability to adapt the functionality it provides to the users. However, the main problems of the systems present in literature are their incapability to identify user needs and, more importantly, the insufficient mappings of those needs to available resources/services. In this paper, we present a two-phase construction of the user model: firstly, an initial static user model is built for the user connecting to the system the first time. Then, the model is revised/adjusted by considering the information collected in the logs of the user interaction with the device/context in order to make the model more adequate to the evolving user’s interests/ preferences/behaviour. The initial model is built by exploiting the stereotype concept, its adjustment is performed exploiting machine learning techniques and particularly, sequence mining and pattern discovery strategies.

Download Full-text