Machine Learning in Local Positioning Systems

Nowadays machine learning (ML) is used in all sorts of fields like health care, retail, travel, finance, social media, etc. ML system is used to learn from input data to construct a suitable model by continuously estimating, optimizing, and tuning parameters of the model. To attain the stated, Python programming language is one of the most flexible languages, and it does contain special libraries for ML applications, namely SciKit-Learn, TensorFlow, PyTorch, Keras, Theano, etc., which is great for linear algebra and getting to know kernel methods of machine learning. The Python programming language is great to use when working with ML algorithms and has easy syntax relatively. When taking the deep-dive into ML, choosing a framework can be daunting. The most common concern is to understand which of these libraries has the most momentum in ML system modeling and development. The major objective of this paper is to provide extensive knowledge on various Python libraries and different ML algorithms in comparison with meet multiple application requirements. This paper also reviewed various ML algorithms and application domains.KeywordsMachine LearningLibrariesPython

The main features of the use of the Python programming language in the field of modeling artificial neural networks are considered. It is shown that the developed system of libraries of the Python programming language significantly simplifies the process of developing neural networks. It is noted that one of the key advantages of the Python language is its convenience for rapid prototyping. At the same time, the interpreted nature of the language allows you to quickly test different architectures of neural networks, change parameters and obtain results in real time. It is noted that the Python programming language provides easy integration with other programming languages and frameworks. It is shown that the TensorFlow and Keras libraries are powerful tools for modeling and training neural networks. They greatly simplify the development process by providing a wide range of features for creating, optimizing, and deploying machine learning models.

<p><span>Image processing is used to classify lung images with malignant or normal nodules. The Convolutional Neural Network (CNN) method is often used to classify images. This study uses a modified CNN architecture with various layers, filters, batch size, dropout, and epoch values. Variations were made to determine the best accuracy value and reduce the overfitting value of the proposed CNN architecture. This study implements the method using the Keras library with the Python programming language. The data is in the form of CT-Scan images of lung cancer and normal lungs. The results of several experiments from the proposed model produce an accuracy value of 95% using three layers, 128 filters on the first layer, 256 on the second layer, and 512 filters on the third layer, then with 32 batch sizes, 0.5 dropout.</span></p>

The Coronavirus is a virus that spreads very quickly. Therefore, it has had very destructive effects in many areas worldwide. Because X-ray images are an easily accessible, fast, and inexpensive method, they are widely used worldwide to diagnose COVID-19. This study tried detecting COVID-19 from X-ray images using pre-trained VGG16, VGG19, InceptionV3, and Resnet50 CNN architectures and modified versions of these architectures. The fully connected layers of the pre-trained architectures have been reorganized in the modified CNN architectures. These architectures were trained on binary and three-class datasets, revealing their classification performance. The data set was collected from four different sources and consisted of 594 COVID-19, 1345 viral pneumonia, and 1341 normal X-ray images. Models are built using Tensorflow and Keras Libraries with Python programming language. Preprocessing was performed on the dataset by applying resizing, normalization, and one hot encoding operation. Model performances were evaluated according to many performance metrics such as recall, specificity, accuracy, precision, F1-score, confusion matrix, ROC analysis, etc., using 5-fold cross-validation. The highest classification performance was obtained in the modified VGG19 model with 99.84 % accuracy for binary classification (COVID-19 vs. Normal) and in the modified VGG16 model with 98.26 % accuracy for triple classification (COVID-19 vs. Pneumonia vs. Normal). These models have a higher accuracy rate than other studies in the literature. In addition, the number of COVID-19 X-ray images in the dataset used in this study is approximately two times higher than in other studies. Since it is obtained from different sources, it is irregular and does not have a standard. Despite this, it is noteworthy that higher classification performance was achieved than in previous studies. Modified VGG16 and VGG19 models (available at github.com/akaraci/LargeDatasetCovid19) can be used as an auxiliary tool in slight healthcare organizations’ shortage of specialists to detect COVID-19.

With the technological development and change of the times in the current era, with the rapid development of science and technology and information technology, there is a gradual replacement in the traditional way of cognition. Effective data analysis is of great help to all societies, thereby drive the development of better interests. How to expand the development of the overall information resources in the process of utilization, establish a mathematical analysis–oriented evidence theory system model, improve the effective utilization of the machine, and achieve the goal of comprehensively predicting the target behavior? The main goal of this article is to use machine learning technology; this article defines the main prediction model by python programming language, analyzes and forecasts the data of previous World Cup, and establishes the analysis and prediction model of football field by K-mean and DPC clustering algorithm. Python programming is used to implement the algorithm. The data of the previous World Cup football matches are selected, and the built model is used for the predictive analysis on the Python platform; the calculation method based on the DPC-K-means algorithm is used to determine the accuracy and probability of the variables through the calculation results, which develops results in specific competitions. Research shows how the machine wins and learns the efficiency of the production process, and the machine learning process, the reliability, and accuracy of the prediction results are improved by more than 55%, which proves that mobile algorithm technology has a high level of predictive analysis on the World Cup football stadium.

With the tremendous increase in the number of songs produced, categorizing them has become a difficult process. It's a good idea to categorize them according to their genre. To resolve this quandary, this paper introduces a web application that uses Machine Learning techniques to automatically classify a music file into its genre. Any .wav audio file can be uploaded to the application, which will accurately assess its genre and classify it accordingly. Django is used to create the web application, and three algorithms are tested for accuracy and performance. Of these three methods, the Support Vector Machine approach has the highest accuracy – Deep Neural Network (Keras library), Support Vector Machine, K Nearest Neighbors – and is thus chosen for the web application.

The object of this study is the process of determining the coordinates of low-visible aerial objects. The main hypothesis of the research assumed that the signals emitted by airborne systems of airborne objects that are not visible to radar stations have a greater power than the signal reflected from the airborne object. This, in turn, could improve the signal/noise ratio and, accordingly, the accuracy of determining the coordinates of low-visible aerial objects. It is suggested to use Software-Defined Radio receivers to receive such signals emitted by on-board systems of low-visible aerial objects. It was established that the main sources of signals for Software-Defined Radio receivers are signals of command, telemetry, target channels, manual control channels, and satellite navigation. It was established that an additional distinguishing feature when determining the coordinates of low-visible aerial objects is the uniqueness of their spectra and spectrograms. The method of determining the coordinates of low-visible aerial objects when using Software-Defined Radio receivers has been improved, which, unlike the known ones, involves: – the use as signals for Software-Defined Radio of signal receivers of on-board equipment of low-visible aerial objects; – the use of a priori coordinate values of a low-visible aerial object; – conducting additional spectral analysis of signals of on-board systems of low-visible aerial objects. The spectra and spectrograms of signals of on-board systems of aerial objects when using non-directional and directional antennas were experimentally determined. The experimental studies confirm the possibility of using the Software-Defined Radio receiver to receive signals from airborne equipment and improve the signal-to-noise ratio. The accuracy of determining the coordinates of aerial objects when using Software-Defined Radio receivers was evaluated. A decrease in the error of determining plane coordinates by the Software-Defined Radio system of receivers compared to the accuracy of determining coordinates by the P-19 MA radar station was established by an average of 1.88–2.47 times, depending on the distance to the aerial object

Currently, in large cities, there is a steady tendency to increase the spatial density of telecommunication systems. The saturation of the radio spectrum with analog and digital systems used to solve the problems of radio communication and television allows to improve the technologies of semi-active radar detection and determination of the coordinates of a low-altitude object on their basis. Conducting radar surveillance using non-radar radar transmitters is often called semi-active radar using extraneous or parasitic radiation sources. The advantages of such systems are the minimization of deployment costs, low operating energy costs, low probability of malfunctioning, stealth of the fact of operation, environmental friendliness, and the absence of requirements for the allocation of radio frequency resources. Relatively high heights of antennas of communications and television transmitters with the existing radiated power create favorable conditions for detecting low-altitude objects. Digital signals of modern telecommunication systems have a spectrum width that provides acceptable resolution and accuracy of measuring the total distance and angular coordinates [20; 21]. In the general case, systems of this type are a multi-position system consisting of one or more radiation sources and one or more receiving positions spread in space [22; 23]. The paper gives a general description of the proposed method of determining the coordinates of an aerial object at low altitude under the conditions of propagation of several beams of radio signals. The method and its technical solution capable of determining the coordinates of an unauthorized low-altitude object under the conditions of the existence of several rays of radio signal propagation are defined, and the algorithm of its operation necessary for the technical implementation of the proposed method is presented. The paper considers options for determining the coordinates of aerial objects with different composition of primary coordinate measurements and the number of receiving points. The accuracy of determining the location of the object for multi-position radio systems of this type was evaluated under the conditions of the existence of several beams of radio signal propagation in the developed model, taking into account the number of received signals and their measurement errors.

Research methods in machine learning play a pivotal role since the accuracy and reliability of the results are influenced by the research methods used. The main aims of this paper were to explore current research methods in machine learning, emerging themes, and the implications of those themes in machine learning research. To achieve this the researchers analyzed a total of 100 articles published since 2021 in IEEE journals. This study revealed that Machine learning uses quantitative research methods with experimental research design being the de facto research approach. The study also revealed that researchers nowadays use more than one algorithm to address a problem. Optimal feature selection has also emerged to be a key thing that researchers are using to optimize the performance of Machine learning algorithms. Confusion matrix and its derivatives are still the main ways used to evaluate the performance of algorithms, although researchers are now also considering the processing time taken by an algorithm to execute. Python programming languages together with its libraries are the most used tools in creating, training, and testing models. The most used algorithms in addressing both classification and prediction problems are; Naïve Bayes, Support Vector Machine, Random Forest, Artificial Neural Networks, and Decision Tree. The recurring themes identified in this study are likely to open new frontiers in Machine learning research. Keywords: Research methods in machine learning, machine learning algorithms, machine learning techniques.

Предметом статті є вивчення можливостей використання технологій MLAT та приймачів ADS-B для підвищення точності визначення координат повітряних об’єктів. Метою є підвищення точності визначення координат повітряних об’єктів за рахунок використання технології MLAT та системи приймачів ADS-B. Завдання: аналіз різницево-далекомірного алгоритму, пропозиції щодо реалізації технології мультилатерації з використанням приймачів ADS-B в інтересах радіотехнічних військ, стислий аналіз можливостей технології ADS-B, аналіз можливостей використання приймачів ADS-B для підвищення точності визначення координат повітряного об’єкта. Використовуваними методами є: методи радіолокації, методи теорії прийому та обробки сигналів, визначення координат повітряних об’єктів. Отримані такі результати. Визначено фактори, що впливають на точність визначення координат повітряних об’єктів. Встановлено можливість застосування технології автоматичного залежного спостереження та технології мультилатерації для підвищення точності визначення координат повітряних об’єктів. Визначено інтервали відхилення часових затримок в залежності від розташування повітряного об’єкта відносно системи приймачів для забезпечення похибки визначення координат, що не перевищуватиме 300 м. Визначена залежність похибки визначення положення повітряного об’єкта від кількості приймачів. Висновки. Наукова новизна отриманих результатів полягає в наступному. Встановлено, що перевагами використання технології MLAT та системи приймачів ADS-B для визначення координат повітряного об’єкта є зменшення похибки визначення положення повітряного об’єкта. Використання технології MLAT та системи приймачів ADS-B не потребує внесення значних змін до існуючих засобів радіолокації. В свою чергу, використання технології MLAT та системи приймачів ADS-B є додатковим джерелом отримання інформації про повітряну обстановку. В подальших дослідженнях пропонується використання технології MLAT та системи приймачів ADS-B на позиціях радіотехнічних підрозділів при веденні радіолокаційного контролю повітряної обстановки

This paper discusses the potential of machine learning, data science, and natural language processing (NLP) in mitigating the incidence of spoofing and financial risks hinged on cyber threats. Another one is spoofing; it is the act of impersonating legitimate entities to gain unauthorized information, and it is indeed a threat to the public and companies to some extent. The research introduces two primary methodologies to combat spoofing: an email filtering system using a machine learning algorithm and an encryption and decryption system using a Caesar Cipher and Python programming language. It distinguishes between approved domains and unapproved domains by using machine learning and successfully filters out phishing emails from reaching the intended clients. This study also illustrates how to conduct email domain verification using MongoDB Atlas, which a database is containing approved vendors’ domains, to reduce spoofing. Specifically, incorporating NLP helps the system analyze raw data to categorize it and identify patterns potentially leading to a spoofing attempt, enhancing the spoofing detection and prevention of the system. The paper also presents arguments that require awareness and integration of new technologies in the security frameworks. Hence, incorporating machine learning, data science, and NLP presents robust, versatile, and cost-effective solutions to enhance cybersecurity and ultimately protect vital information and organizations’ monetary loss due to cybercrimes. The paper was first completed in 2021 and later I modified the article with latest updates till date 2024.

The “youngest” of the major satellite applications is the field of satellite navigation. This field of measurement and ranging through the use of satellite positioning systems first started in the context of scientific research. This initial use of satellites was simply for positioning and location. These activities, that were first based on using Doppler frequency shifts as a satellite orbited above, were largely scientific and not strategic. These types of activities included geodetics (i.e., such as the measurement of continental drift over time) or the collection of scientific information and data from atmospheric landor ocean-based sensors where the specific locations of the sensors were important. The real strides in the development of satellite navigation, however, came when space systems were developed for the specific purpose of precise targeting of missiles and various other types of weapon systems. The Soviet Union/Russian J.N. Pelton (*) International Space University, Arlington, VA, USA e-mail: joepelton@verizon.net S. Camacho-Lara Centro Regional de Ensenanza de Ciencia y Tecnologia del Espacio para America Latina y el Caribe (CRECTEALC), Santa Maria Tonantzintla, Puebla, Mexico e-mail: sergio.camacho@inaoep.mx # Springer Science+Business Media New York 2016 J.N. Pelton et al. (eds.), Handbook of Satellite Applications, DOI 10.1007/978-1-4614-6423-5_10-3 1 Federation satellite navigation system, known as GLONASS, and the US-based Navstar system –with its Global Positioning Satellite (GPS) network – developed space-based systems that provided unprecedented capabilities to determine locations on the ground or oceans with great accuracy. Early systems such as Argos that relied on Doppler shift technology were only accurate within a precision of hundreds of meters. Today’s advanced satellite navigation systems, however, are accurate for measurements that can be indicated with a precision only a few meters and within centimeters if utilizing reference stations. The fact that GPS and GLONASS satellite signals are freely available in space for all to use has spurred the development of low-cost receivers for much more than strategic or military usage. Today, various types of civilian use of navigational and positioning satellites have become popular on a global basis. The latest development in application-specific integrated circuit (ASIC) chip technology has fueled the growth of applications based on the use of these precise spacebased navigational and positioning systems. The availability of these highly capable but increasingly low-cost chips – small enough that they could be included in handsets such as cell phones or included with the electronics available on various types of vehicles (i.e., cars, trucks, buses, trains, airplanes, ships, etc.) – represented a real breakthrough. The development of specialized computer chips to perform satellite navigation calculations has increased the number of users of this technology from a relatively small population of several thousands to tens of millions. In the coming decade, the proliferation of international satellite navigation systems in orbit (i.e., USA, Russian Federation, China, India, Europe, and Japan) plus the continued development of ever lower cost of satellite navigation chips for receivers will continue the popular expansion of these increasingly “easy to use” and versatile systems. Wide and easy access to low-cost consumer receivers for satellite navigation services strongly suggests that this trend of expanded use will continue to surge. Thus, within a decade, use of these devices will increase to the hundreds of millions if not billions of people. Atomic clocks with incredible temporal precision today allow satellite navigation systems to be used by military organizations, governmental geospatial scientists, and scientists to determine locations with great accuracy. But these applications are now just a small fraction of total usage. The deployment of the new international satellite navigational systems and smaller and lower cost ground receivers will support an ever-increasing civilian consumer market for an ever-expanding range of everyday applications. These consumer applications include driving a vehicle to a desired location, safely sailing a boat, going mountain climbing or hiking without getting lost, or simply finding out where you are within a city. The precise time keeping ability of today’s satellite navigation systems also means that these spacecrafts can also serve as a global timekeeper for computers and scientific experiments. The time stamp from a satellite navigation satellite can also be used not only for scientific or regulatory purposes but for other applications such as security and banking systems as well. These satellites are also used to support the synchronization needs of communications satellites. In short, 2 J.N. Pelton and S. Camacho-Lara

Several studies applying Machine Learning to deception detection have been published in the last decade. A rich and complex set of settings, approaches, theories, and results is now available. Therefore, one may find it difficult to identify trends, successful paths, gaps, and opportunities for contribution. The present literature review aims to provide the state of research regarding deception detection with Machine Learning. We followed the PRISMA protocol and retrieved 648 articles from ACM Digital Library, IEEE Xplore, Scopus, and Web of Science. 540 of them were screened (108 were duplicates). A final corpus of 81 documents has been summarized as mind maps. Metadata was extracted and has been encoded as Python dictionaries to support a statistical analysis scripted in Python programming language, and available as a collection of Jupyter Lab Notebooks in a GitHub repository. All are available as Jupyter Lab Notebooks. Neural Networks, Support Vector Machines, Random Forest, Decision Tree and K-nearest Neighbor are the five most explored techniques. The studies report a detection performance ranging from 51% to 100%, with 19 works reaching accuracy rate above 0.9. Monomodal, Bimodal, and Multimodal approaches were exploited and achieved various accuracy levels for detection. Bimodal and Multimodal approaches have become a trend over Monomodal ones, although there are high-performance examples of the latter. Studies that exploit language and linguistic features, 75% are dedicated to English. The findings include observations of the following: language and culture, emotional features, psychological traits, cognitive load, facial cues, complexity, performance, and Machine Learning topics. We also present a dataset benchmark. Main conclusions are that labeled datasets from real-life data are scarce. Also, there is still room for new approaches for deception detection with Machine Learning, especially if focused on languages and cultures other than English-based. Further research would greatly contribute by providing new labeled and multimodal datasets for deception detection, both for English and other languages.

Research methods in machine learning play a pivotal role since the accuracy and reliability of the results are influenced by the research methods used. The main aims of this paper were to explore current research methods in machine learning, emerging themes, and the implications of those themes in machine learning research. To achieve this the researchers analyzed a total of 100 articles published since 2019 in IEEE journals. This study revealed that Machine learning uses quantitative research methods with experimental research design being the de facto research approach. The study also revealed that researchers nowadays use more than one algorithm to address a problem. Optimal feature selection has also emerged to be a key thing that researchers are using to optimize the performance of Machine learning algorithms. Confusion matrix and its derivatives are still the main ways used to evaluate the performance of algorithms, although researchers are now also considering the processing time taken by an algorithm to execute. Python programming languages together with its libraries are the most used tools in creating, training, and testing models. The most used algorithms in addressing both classification and prediction problems are; Naïve Bayes, Support Vector Machine, Random Forest, Artificial Neural Networks, and Decision Tree. The recurring themes identified in this study are likely to open new frontiers in Machine learning research.

Machine learning (ML) is becoming increasingly important as a research tool due to its various frameworks and learning approaches. With the ever-increasing scale of software, reliability has become a crucial issue and software defect prediction is utilized to assist developers in finding potential defect and allocating their testing efforts. Traditional methods of software defect prediction mainly focus on designing static code metrics which are fed into ML classifiers to predict defects in the code. Even with the same ML techniques, many researchers apply statistical approaches to classify software modules and decide whether each module is defect prone or not and, accordingly, train their model. Deep neural network (DNN) and convolutional neural network (CNN) models built by the appropriate design decisions are crucial to obtain the desired classifier performance. This is especially significant when predicting fault proneness of software modules. When correctly identified, this could help in reducing the testing cost by directing the efforts more toward the modules identified to be fault prone. This paper proposed a N ovel CNN (NCNN) model to predict software defects. The framework used is Python Programming Language with Keras and TensorFlow. A comparative analysis with ML algorithms [such as Random Forest (RF), Decision Trees (DT), and Naïve Bayes (NB)] and DNN model in terms of F-measure (known as F1-score), recall, precision, and accuracy has been presented from four NASA system data sets (KC1, PC1, PC2, and KC3) selected from PROMISE repository. The experimental results indicated that NCNN model was comparable to the existing classifiers and outperformed them in most of the experiments.