Fake News Detection Using Machine Learning Ensemble Methods

The World Wide Web's introduction, as well as the rapid adoption of social media platforms (such as Facebook and Twitter), paved the way for unprecedented levels of knowledge distribution in human history. Certain entities, however, take advantage of such platforms, sometimes for monetary gain and sometimes for the purpose of promoting prejudiced viewpoints, changing mindsets, and disseminating satire or ridiculousness. This phenomenon is known as fake news. Because the spread of fake news can have serious consequences, such as election rigging and widening political divides, developing methods to detect fake news content is critical. A computerised classification system is used to classify a wide range of things. Fake news is a term used to describe this occurrence. Given the serious consequences of fake news, such as swaying elections and widening political divides, developing methods for detecting fake news content is critical. It is difficult to automate the classification of a text article as misinformation or disinformation. We suggest employing a machine learning ensemble approach for automatic classification in this paper. Our research looks into various textual properties such as word counts, term frequency, and inverse document frequency that can be used to distinguish between fake and real content. We train a Passive-Aggressive Classifier and evaluate its performance on real-world datasets using these properties. The datasets will be used in part to train and part to test the classifiers. It is possible to determine whether the news is fake or not by testing the different features of the datasets on the classifier.

Electron stopping power (SP) is of great importance in theoretical and applied research areas specifically for Monte Carlo simulation studies in many microanalysis and surface analysis techniques, radiation dosimetry, and the design of particle detectors. However, experimental data are available for a dozen elemental materials only. On the other hand, the Bethe analytical expression of the SP is applicable at high energies only whereas no generally accepted formula exists at lower energies. We employed ensemble machine learning (ML) methods with the available experimental database for the prediction of SPs of electrons with energies from 100 keV down to 1 eV, in elements over the entire periodic table. With a small training database for electron SPs, we applied various algorithms individually as well as their ensembles, which have the credibility to enhance the prediction accuracy in the case of a small training database. Based on the model's performance evaluation tests, we concluded that the stacked generalization is more accurate than the individual algorithms. Using this method, we were able to predict the electron SPs for 54 elements (in total) including 12 elements that were present in the training database as well as for 42 elements beyond the training database over a wide energy range (1 eV to 100 keV). Compared to other theoretical approaches, the ML predicted SPs show very good agreement with the available experimental data at all energies. Moreover, unlike other theoretical approaches, the ML model does not need dielectric function data and other physical parameters which involve complex calculations. Using our ML model, we have predicted SPs for a further 14 elements for which no theoretical SPs are available because of the lack of good dielectric function data.

Behind the cues: A benchmarking study for fake news detection

Estimation of time dependent scour depth around circular bridge piers: Application of ensemble machine learning methods

Landslide disaster risk reduction necessitates the investigation of different geotechnical causal factors for slope failures. Machine learning (ML) techniques have been proposed to study causal factors across many application areas. However, the development of ensemble ML techniques for identifying the geotechnical causal factors for slope failures and their subsequent prediction has lacked in literature. The primary goal of this research is to develop and evaluate novel feature selection methods for identifying causal factors for slope failures and assess the potential of ensemble and individual ML techniques for slope failure prediction. Twenty-one geotechnical causal factors were obtained from 60 sites (both landslide and non-landslide) spread across a landslide-prone area in Mandi, India. Relevant causal factors were evaluated by developing a novel ensemble feature selection method that involved an average of different individual feature selection methods like correlation, information-gain, gain-ratio, OneR, and F-ratio. Furthermore, different ensemble ML techniques (Random Forest (RF), AdaBoost (AB), Bagging, Stacking, and Voting) and individual ML techniques (Bayesian network (BN), decision tree (DT), multilayer perceptron (MLP), and support vector machine (SVM)) were calibrated to 70% of the locations and tested on 30% of the sites. The ensemble feature selection method yielded six major contributing parameters to slope failures: relative compaction, porosity, saturated permeability, slope angle, angle of the internal friction, and in-situ moisture content. Furthermore, the ensemble RF and AB techniques performed the best compared to other ensemble and individual ML techniques on test data. The present study discusses the implications of different causal factors for slope failure prediction.

Predicting energy consumption helps countries make strategic decisions in many critical areas such as energy management, economic development, energy security, environmental sustainability and infrastructure investments. Therefore, accurate and reliable energy consumption predictions are vital to ensure the sustainability and prosperity of countries. This study aims to contribute to the proper planning of transportation policies and energy management by successfully predicting Türkiye's railway energy consumption. In this direction, energy prediction values were obtained from 18 different machine learning methods using the country's railway line length, number of passengers, freight amount and energy consumption values from 1977 to 2024. To further strengthen the results obtained with these methods, bagging, boosting, stacking and blending ensemble learning methods were utilized. With the improvements, the R‐squared value was increased up to 0.9667 and energy predicting was achieved with very high accuracy. Based on the results obtained from this study, it is possible to provide investment planning more efficiently. In addition, the implementation of energy management strategies, infrastructure planning and sustainable energy policies will be provided more efficiently as a result of obtaining more successful results by using ensemble machine learning methods instead of traditional machine learning methods for energy consumption predictions in different sectors.

The accuracy of electric load forecasting is crucial when working on applications in power grid decision-making and operation. Due to the non-linear and stochastic behaviour of customers, the electric load profile is a complicated signal. In this paper, authors propose machine learning based automated system for electricity load forecasting, taking into consideration various variable factors that have an impact on the future electricity load demand. Three machine learning algorithms are used for evaluation of the proposed framework. The algorithms are evaluated on electricity load data collected from eastern region of Ontario, integrated with the weather and population data of the region. The Light GBM algorithm comparatively performs best with mean absolute error of 0.156. The developed system can be used for more accurate and efficient load forecasting applications.

The remarkable advances in ensemble machine learning methods have led to a significant analysis in large data, such as random forest algorithms. However, the algorithms only use the current features during the process of learning, which caused the initial upper accuracy’s limit no matter how well the algorithms are. Moreover, the low classification accuracy happened especially when one type of observation’s proportion is much lower than the other types in training datasets. The aim of the present study is to design a hierarchical classifier which try to extract new features by ensemble machine learning regressors and statistical methods inside the whole machine learning process. In stage 1, all the categorical variables will be characterized by random forest algorithm to create a new variable through regression analysis while the numerical variables left will serve as the sample of factor analysis (FA) process to calculate the factors value of each observation. Then, all the features will be learned by random forest classifier in stage 2. Diversified datasets consist of categorical and numerical variables will be used in the method. The experiment results show that the classification accuracy increased by 8.61%. Meanwhile, it also improves the classification accuracy of observations with low proportion in the training dataset significantly.

Fake news analysis and detection is a method that requires using machine learning classifiers to recognize and detect news content as fake or real. No research work has been carried out on fake news analysis and detection using the machine learning algorithms using Truncated singular value decomposition (TSVD) on the ‘ISOT Fake News Dataset’ but there is a need for a better design that can provide better accuracy. This research aims to develop and use systematic progression in identifying fake news analysis using machine learning algorithms with TSVD. The machine learning algorithms which include the light Gradient Boosting Machine, Cat boosting classifier, and support vector machine with TSVD were used on ‘The ISOT Fake News Dataset’ to recognize the fake news from real news in the dataset, thereby improving the performance of these classifiers by cleaning the data to manage the data imbalances more efficiently. The recommended design yields an exactness of 97.33%. The postulated technique out classifies the current classifiers which possess an exactness of 95.05%. We recommend that Machine learning classifiers using TSVD should be applied in Fake News analysis and detection since it gives a better performance accuracy in fake news analysis and detection

P125. Development of a novel ensemble machine learning algorithm for prediction of complications and readmission after anterior cervical spinal fusion

Cross-cultural gender differences in the adoption and usage of social media platforms – An exploratory study of Last.FM

Recent advances in the machine learning field have given rise to efficient ensemble methods that accurately forecast time-series. In this paper, we use the Quantopian algorithmic stock market trading simulator to assess ensemble methods performance in daily prediction and trading. The ensemble methods used are Extremely Randomized Trees, Random Forest, and Gradient Boosting. All methods are trained using multiple technical indicators and automatic stock selection is used. Simulation results show significant returns relative to the benchmark and large values of alpha are produced from all methods. These results strengthen the role of ensemble method based machine learning in automated stock market trading.

This paper analyses the recent evolution of the electricity price of one of the EastEuropean countries’ Balancing Markets (BM) – Romania, aiming to understand the prices trend and predict them in the current economic and geopolitical context. This is especially important as the electricity producers have to allocate their output between wholesale electricity market, ancillary services markets and BM targeting to maximize value and achieve a sustainable economic development. Therefore, in this paper, we propose an AI-powered electricity price forecast using several types of standout Machine Learning (ML) algorithms such as classifiers and regressors to predict the electricity price on BM. This approach, consisting of two steps, identifies the imbalance sign and significantly enhances the performance of the price forecast. The proposed method offers valuable insights into the market participants’ trading opportunities using two prediction solutions. The first prediction solution consists of averaging the results of five ensemble ML algorithms. The second one consists in weighting the results of the five forecasting ML algorithms using either a linear regression or a decision tree algorithm. Thus, we propose to combine supervised and unsupervised ML algorithms and find the fundamentals for creating optimal bidding strategies for electricity market players.

Social media have diffused into the everyday lives of many but still pose challenges to individuals regarding use of these platforms. This paper explores the multiple manners in which social media platforms gets employed by individuals based on an examination of 4 vignettes generated by interviewing individuals within a university context. An analysis of the vignettes and individual use behaviors highlights the tension between network-based adoption of social media platforms and the constraints that the network places on individual use of the platform.Keywordssocial mediaadoptionusesocial networktensions

This study examined the adoption of social media platforms for digital reference service delivery by librarians in federal university libraries in Southeast, Nigeria. The study employed a correlational and descriptive survey design. The total population consisted of 184 librarians in federal university libraries in Southeast, Nigeria total enumeration sampling technique was used to include all 184 librarians in the study. Data collection was conducted using a questionnaire, which was validated by senior lecturers in the Department of Library and Information Science at Ignatius Ajuru University of Education in the College of Education, Warri Centre. The reliability of the questionnaire was assessed using Pearson’s Product-Moment Correlation coefficient, resulting in the following reliability indices: Benefits Scale = 0.92; Willingness Scale = 0.83; and Readiness Scale = 0.70. With all indices exceeding 0.60, the instrument was deemed reliable. Data collected were analyzed with mean and standard deviation to answer the research questions, with a weighted mean of 2.50 used as the decision threshold. The hypothesis was tested using Pearson Product Moment Correlation (PPMC), at a 0.05 level of significance. Out of 184 distributed questionnaires, 110 (60%) were returned and analyzed. The findings revealed that librarians demonstrated a high level of readiness and willingness to adopt social media platforms for digital reference service delivery. They also perceived numerous benefits of using these platforms, including improved communication and service delivery. The study further revealed significant relationships between librarians’ adoption of social media and willingness to utilize these platforms for digital reference services. Additionally, a moderate correlation was found among adoption, and willingness to adopt social media platforms for digital reference service delivery. In conclusion, the study established that librarians in federal university libraries in Southeast, Nigeria are highly willing to adopt social media for digital reference services. It was recommended that library management in these institutions should support the professional development of librarians by organizing workshops and training programmes to enhance their skills in utilizing social media platforms for digital reference services.