Averaged Perceptron Research Articles

Application of interval type-2 fuzzy logic and type-1 fuzzy logic-based approaches to social networks for spam detection with combined feature capabilities.

Social networks are large platforms that allow their users to interact with each other on the Internet. Today, the widespread use of social networks has made them vulnerable to malicious use through different methods such as fake accounts and spam. As a result, many social network users are exposed to the harmful effects of spam accounts created by malicious people. Although Twitter, one of the most popular social networking platforms, uses spam filters to protect its users from the harmful effects of spam, these filters are insufficient to detect spam accounts that exhibit new methods and behaviours. That's why on social networking platforms like Twitter, it has become a necessity to use robust and more dynamic methods to detect spam accounts. Fuzzy logic (FL) based approaches, as they are the models such that generate results by interpreting the data obtained based on heuristics viewpoint according to past experiences, they can provide robust and dynamic solutions in spam detection, as in many application areas. For this purpose, a data set was created by collecting data on the twitter platform for spam detection. In the study, fuzzy logic-based classification approaches are suggested for spam detection. In the first stage of the proposed method, a data set with extracted attributes was obtained by applying normalization and crowdsourcing approaches to the raw data obtained from Twitter. In the next stage, as a process of the data preprocessing step, six attributes in the binary form in the data set were subjected to a rating-based transformation and combined with the other real-valued attribute to create a database to be used in spam detection. Classification process inputs were obtained by applying the fisher-score method, one of the commonly used filter-based methods, to the data set obtained in the second stage. In the last stage, the data were classified based on FL based approaches according to the obtained inputs. As FL approaches, four different Mamdani and Sugeno fuzzy inference systems based on interval type-1 and Interval Type-2 were used. Finally, in the classification phase, four different machine learning (ML) approaches including support vector machine (SVM), Bayesian point machine (BPM), logistic regression (LR) and average perceptron (Avr Prc) methods were used to test the effectiveness of these approaches in detecting spam. Experimental results were obtained by applying different FL and ML based approaches on the data set created in the study. As a result of the experiments, the Interval Type-2 Mamdani fuzzy inference system (IT2M-FIS) provided the highest performance with an accuracy of 0.955, a recall of 0.967, an F-score 0.962 and an area under the curve (AUC) of 0.971. However, it has been observed that FL-based spam models have a higher performance than ML-based spam models in terms of metrics including accuracy, recall, F-score and AUC values.

Autoregressive Integrated Moving Average vs. Artificial Neural Network in Predicting COVID-19 Cases in Malaysia

On March 11,2020, the World Health Organization (WHO) declared Covid-19 as a global pandemic. The spread of Covid-19 has threatened many lives in nearly every country. In Malaysia, the health authorities have expressed concerns over an increasing number of cases and deaths. Due to the lockdown, this pandemic has also had an impact on most economic activities. Consequently, it is crucial to develop a reliable forecasting model to anticipate the number of cases. This study proposes two models: Autoregressive Integrated Moving Average (ARIMA) and Multilayer Perceptron Neural Network (MPNN) in predicting the number of Covid-19 cases in Malaysia. Using Mean Absolute Error (MAE), the effectiveness and forecasting accuracy of the two models are compared and assessed. The lowest the value of MAE, the more accurate the forecasted outputs. The secondary data used in this study was the average number of Covid-19 cases each day in Malaysia from March 1, 2020, to March 29, 2021. To evaluate the data, RStudio and Alyuda NeuroIntelligence are utilised. As a consequence, the ARIMA (4,1,5) model provided the best fit to the data when compared to other ARIMA models, with a Mean Absolute Error (MAE) score of 1096.799. However, Multilayer Perceptron Neural Network (MPNN), which had the lowest MAE value of 334.591, outperformed ARIMA in terms of performance. The MPNN model was then used to forecast the number of Covid-19 instances for the next 30 days. According to the findings, daily increases in cases are anticipated.

Modeling of Groundwater Potential Using Cloud Computing Platform: A Case Study from Nineveh Plain, Northern Iraq

Knowledge of the groundwater potential, especially in an arid region, can play a major role in planning the sustainable management of groundwater resources. In this study, nine machine learning (ML) algorithms—namely, Artificial Neural Network (ANN), Decision Jungle (DJ), Averaged Perceptron (AP), Bayes Point Machine (BPM), Decision Forest (DF), Locally-Deep Support Vector Machine (LD-SVM), Boosted Decision Tree (BDT), Logistic Regression (LG), and Support Vector Machine (SVM)—were run on the Microsoft Azure cloud computing platform to model the groundwater potential. We investigated the relationship between 512 operating boreholes with a specified specific capacity and 14 groundwater-influencing occurrence factors. The unconfined aquifer in the Nineveh plain, Mosul Governorate, northern Iraq, was used as a case study. The groundwater-influencing factors used included elevation, slope, curvature, topographic wetness index, stream power index, soil, land use/land cover (LULC), geology, drainage density, aquifer saturated thickness, aquifer hydraulic conductivity, aquifer specific yield, depth to groundwater, distance to faults, and fault density. Analysis of the contribution of these factors in groundwater potential using information gain ratio indicated that aquifer saturated thickness, rainfall, hydraulic conductivity, depth to groundwater, specific yield, and elevation were the most important factors (average merit > 0.1), followed by geology, fault density, drainage density, soil, LULC, and distance to faults (average merit < 0.1). The average merits for the remaining factors were zero, and thus, these factors were removed from the analysis. When the selected ML classifiers were used to estimate groundwater potential in the Azure cloud computing environment, the DJ and BDT models performed the best in terms of all statistical error measures used (accuracy, precision, recall, F-score, and area under the receiver operating characteristics curve), followed by DF and LD-SVM. The probability of groundwater potential from these algorithms was mapped and visualized into five groundwater potential zones: very low, low, moderate, high, and very high, which correspond to the northern (very low to low), southern (moderate), and middle (high to very high) portions of the study area. Using a cloud computing service provides an improved platform for quickly and cheaply running and testing different algorithms for predicting groundwater potential.

Modeling of a simplified hybrid algorithm for short-term load forecasting in a power system network

PurposeThe purpose of this paper is to develop a hybrid algorithm, which is a blend of auto-regressive integral moving average (ARIMA) and multilayer perceptron (MLP) for addressing the non-linearity of the load time series.Design/methodology/approachShort-term load forecasting is a complex process as the nature of the load-time series data is highly nonlinear. So, only ARIMA-based load forecasting will not provide accurate results. Hence, ARIMA is combined with MLP, a deep learning approach that models the resultant data from ARIMA and processes them further for Modelling the non-linearity.FindingsThe proposed hybrid approach detects the residuals of the ARIMA, a linear statistical technique and models these residuals with MLP neural network. As the non-linearity of the load time series is approximated in this error modeling process, the proposed approach produces accurate forecasting results of the hourly loads.Originality/valueThe effectiveness of the proposed approach is tested in the laboratory with the real load data of a metropolitan city from South India. The performance of the proposed hybrid approach is compared with the conventional methods based on the metrics such as mean absolute percentage error and root mean square error. The comparative results show that the proposed prediction strategy outperforms the other hybrid methods in terms of accuracy.

Forecasting of Tomato Yields Using Attention-Based LSTM Network and ARMA Model

Nonlinear autoregressive exogenous (NARX), autoregressive integrated moving average (ARIMA) and multi-layer perceptron (MLP) networks have been widely used to predict the appearance value of future points for time series data. However, in recent years, new approaches to predict time series data based on various networks of deep learning have been proposed. In this paper, we tried to predict how various environmental factors with time series information affect the yields of tomatoes by combining a traditional statistical time series model and a deep learning model. In the first half of the proposed model, we used an encoding attention-based long short-term memory (LSTM) network to identify environmental variables that affect the time series data for tomatoes yields. In the second half of the proposed model, we used the ARMA model as a statistical time series analysis model to improve the difference between the actual yields and the predicted yields given by the attention-based LSTM network at the first half of the proposed model. Next, we predicted the yields of tomatoes in the future based on the measured values of environmental variables given during the observed period using a model built by integrating the two models. Finally, the proposed model was applied to determine which environmental factors affect tomato production, and at the same time, an experiment was conducted to investigate how well the yields of tomatoes could be predicted. From the results of the experiments, it was found that the proposed method predicts the response value using exogenous variables more efficiently and better than the existing models. In addition, we found that the environmental factors that greatly affect the yields of tomatoes are internal temperature, internal humidity, and CO2 level.

Prediction of the mortality rate and framework for remote monitoring of pregnant women based on IoT

To save the risk during the pregnancies in remote areas (where women cannot approach the doctors in the urban areas for proper check-ups), the authors have proposed an IoT based remote monitoring of pregnant ladies where the data is collected at the cloud server. Machine learning techniques will be applied on the trimesters’ attributes to find out the reasons of mortality rate of the babies. The use of these advanced technologies in pregnant women care environment can absolutely eradicate the pregnancy complications and harmful incidents. An initial work towards this study is to assess mortality risk prediction in pregnant ladies using machine-learning algorithms for proper prediction and treatment on time. A dataset of 10,000 pregnant women is analysed in this study. Classification algorithms are used to check the death rate of new born babies based on the mother’s age. The survival ratio is presented by applying the various algorithms. Two class SVM model is presented as the most accurate prediction model which outperformed over boosted decision tree, average perceptron, decision forest, locally deep SVM, Bayes point machine, decision forest and logistic regression. Age of the mother and frequency of the pregnancy without proper gap (i.e. less than 3 years) make impact on the mortality rate of the babies. Among the various algorithms, the AUC value of decision forest is augmented at 91.4% whereas two class SVM shows significantly improved performance to 96.6%. The authors have proposed machine learning-based final model, death rate of infants, frequency of pregnancy and age of the mother were interrelated as notable risk factors for mortality in Indian ladies along with other issues.

A smart approach for fire prediction under uncertain conditions using machine learning

One of the most ubiquitous cause of worldwide deforestation and devastation of wildlife is fire. To control fire and reach the forest area in time is not always possible. Consequently, the level of destruction is often high. Therefore, predicting fires well in time and taking immediate action is of utmost importance. However, traditional fire prediction approaches often fail to detect fire in time. Therefore, a more reliable approach like the Internet of Things (IoT) needs to be adopted. IoT sensors can not only observe the real-time conditions of an area, but it can also predict fire when combined with Machine learning. This paper provides an insight into the use of Machine Learning models towards the occurrence of forest fires. In this context, eight Machine Learning algorithms: Boosted Decision Trees, Decision Forest Classifier, Decision Jungle Classifier, Averaged Perceptron, 2-Class Bayes Point Machine, Local Deep Support Vector Machine (SVM), Logistic Regression and Binary Neural Network model have been implemented. Results suggest that the Boosted decision tree model with the Area Under Curve (AUC) value of 0.78 is the most suitable candidate for a fire prediction model. Based on the results, we propose a novel IoT-based smart Fire prediction system that would consider both meteorological data and images for early fire prediction.

Parallel hybridization of series (PHOS) models for time series forecasting

Literature indicates that many efforts have been conducted toward the development of forecasting models with a high degree of accuracy. Combining different models is known as a powerful alternative to access more reliable and more accurate results than single models. Given the great importance of hybridization theory, various hybrid models have been proposed in the literature of time series forecasting. Series hybrid approaches are one of the most well study and the most widely used hybridization models, in which components are sequentially applied. However, according to the modeling essence of this hybrid structure, the performance of the series hybrid models is directly dependent on the order of using components and modeling sequence. Besides, selecting the best modeling order that can yield the best performance in all situations is a problematic theoretical as well as practical task. Thus, the main purpose of this paper is to eliminate the drawback of series models, regarding modeling order selection using a parallel hybridization schema, which is addressed for the first time. The core principle of the proposed parallel hybridization of series (PHOS) models is to improve the series hybrid model’s forecasting accuracy by integrating two hybrid structures in contrast to existing hybrid models, which emphasize only the combination of individual models. The proposed model decomposes the original time series into two linear and nonlinear parts and uses the autoregressive integrated moving average (ARIMA) and multilayer perceptron neural network (MLP) models to model underlying patterns, incorporating two series models including ARIMA–MLP and MLP–ARIMA. Finally, the series models are integrated as components of the parallel hybridization scheme. Moreover, an ordinary least square algorithm is developed to determine the optimal weights of these two components. Three benchmark data sets, including the closing of the DAX index, the closing of the Nikkei 225 index (N225), and the opening of the Dow Jones Industrial Average Index, are used for empirical analysis and verifying the effectiveness of the PHOS model. The empirical analysis indicates that the PHOS model can improve the forecasting performance of both series ARIMA–MLP and MLP–ARIMA models as well as individual models and some parallel hybrid models.

Wireless backhaul network’s capacity optimization using time series forecasting approach

The extensive technological participation in our daily life and business world has transformed the technology model predominantly from last decade and it has created many directions for further research and development. In previous generations of technology, the major development was hardware side but now it is shifting to the smart and intelligent solutions based on AI. The new challenges have provoked and complexity is increasing for resource orchestration because of the nonstop increasing network and its applications. In cellular architecture, the backhaul network (intermediate network) is used to connect the radio and core network. The key technology is the point to point fixed links that is used in the wireless backhaul network. More than 60% radio base station are connected using this technology. Currently, the static approach is used for planning and optimization of the network which is now becoming very difficult as the network is increasing continuously and becoming more complex. The dynamic resource allocation method is proposed for the future capacity forecasting system which is founded on the factual employment of point to point links. Capacity is a crucial factor in wireless backhaul network so best capacity optimization can lead to a good frequency reuse and optimal use of other network resources. The development is based on three different models namely (1) Autoregressive (AR), (2) Seasonal Autoregressive Integrated Moving Average (SARIMA) and (3) Multi-Layer Perceptron (MLP) neural network. Root Means Squared Error (RMSE) and Means Absolute Percentage Error (MAPE) are performance criterion that are used to evaluate the models. When we compare the models' performance, the MLP results are most credible but it takes more time to converge than AR and SARIMA. By using the proposed estimation method, the static optimization will positively move to the dynamic (forecasted) optimization and the distribution of capacity utilization will be right skewed. Hence, the proposed system is efficient and has the ability to optimize the network according to the actual network’s capacity utilization. It will assist the network planner to perform more efficiently, resource distribution will be more balanced and the wastage of resources can be reduced.

Statistical and Machine Learning Methods for Vaccine Demand Forecasting: A Comparative Analysis

This study aimed to find a suitable model for forecasting the appropriate stock of vaccines to avoid shortage and over-supply. The Auto-Regressive Integrated Moving Average (ARIMA) and Multilayer Perceptron Neural Network (MLPNN) models were used for forecasting time series data. The monthly vaccination coverage was used to develop the models from January 2014 until December 2019. The dataset consists of 72 months of observation, the 60 months of data are used for model fitting from January 2014 to December 2019, and the remaining 12 months of data from January 2019 to December 2019 are used to test the accuracy of the forecast. The most suitable forecast model was selected based on the lowest Root Mean Square Error (RMSE) value and the Mean Absolute Error (MAE). The analytical result shows that the MLPNN model outperformed the ARIMA model in forecasting monthly demand for vaccines. The results will help policymakers improve the proper use of vaccination resources.

Solar irradiation prediction with machine learning: Forecasting models selection method depending on weather variability

Eleven statistical and machine learning tools are analyzed and applied to hourly solar irradiation forecasting for time horizon from 1 to 6 h. A methodology is presented to select the best and most reliable forecasting model according to the meteorological variability of the site. To make the conclusions more universal, solar data collected in three sites with low, medium and high meteorological variabilities are used: Ajaccio, Tilos and Odeillo. The datasets variability is evaluated using the mean absolute log return value. The models were compared in term of normalized root mean square error, mean absolute error and skill score. The most efficient models are selected for each variability and temporal horizon: for the weak variability, auto-regressive moving average and multi-layer perceptron are the most efficient, for a medium variability, auto-regressive moving average and bagged regression tree are the best predictors and for a high one, only more complex methods can be used efficiently, bagged regression tree and the random forest approach.

Kernel Oriented Multivariate Feature Selection for Breast Cancer Data Classification via MRMR Filter

A feature selection technique is highly preferred preceding data classification to improve prediction performance especially in the high dimensional space. In general, filter techniques can be considered as essential or assistant selection system on account of their effortlessness, adaptability, and low computational many-sided quality. Nonetheless, a progression of inconsequential cases demonstrates that filter techniques result in less precise execution since they disregard the conditions of features. Albeit few publications have committed their regard for uncover the relationship of features by multivariate-based techniques, these strategies depict connections among elements just by linear techniques. While straightforward linear combination relationship limits the transformation in execution. In this paper, we utilized kernel method for svm-RFE with MRMR way to deal with find inalienable nonlinear connections among features and also amongst feature and target. So as to uncover the viability of our technique we played out a few analyses and thought about the outcomes between our technique and other aggressive multivariatebased features selectors. In our examination, we utilized three classifiers (support vector machine, neural system and average perceptron) on two gathering datasets, to be specific two-class and multi-class datasets (principally focused on svm). Exploratory results show that the execution of our technique is superior to anything others, particularly on three hard group datasets, to be specific Wang’s Breast Cancer, Gordon’s Lung Adenocarcinoma and Pomeroy’s Medulloblastoma. Note: Entire Implementation was developed with MS Machine learning studio.

FinnPos: an open-source morphological tagging and lemmatization toolkit for Finnish

This paper describes FinnPos, an open-source morphological tagging and lemmatization toolkit for Finnish. The morphological tagging model is based on the averaged structured perceptron classifier. Given training data, new taggers are estimated in a computationally efficient manner using a combination of beam search and model cascade. The lemmatization is performed employing a combination of a rule-based morphological analyzer, OMorFi, and a data-driven lemmatization model. The toolkit is readily applicable for tagging and lemmatization of running text with models learned from the recently published Finnish Turku Dependency Treebank and FinnTreeBank. Empirical evaluation on these corpora shows that FinnPos performs favorably compared to reference systems in terms of tagging and lemmatization accuracy. In addition, we demonstrate that our system is highly competitive with regard to computational efficiency of learning new models and assigning analyses to novel sentences.

КОМПЛЕКСНА МОДЕЛЬ ПРОГНОЗУВАННЯ СТІЙКОСТІ ФІНАНСОВО-ЕКОНОМІЧНИХ УСТАНОВ

The paper outlines the key principles of an integrated method of analysis of financial stability of the financial and economic organizations of the banking sector. The research focuses on the use of the openly available financial information of banks and recommends a wide range of interested users to make different investment decisions. The article analyzes the mathematical foundations of forecasting the financial stability of banks. The necessity of primary processing of rated incoming information in the form of time series is established. The essence of methodologies of integrated autoregression with moving average and multilayer perceptron neural networks is explained. Provides the recommendations for improving the methods of forecasting financial stability of the bank. The article proves the feasibility of using a comprehensive application of the modelling methodology. The results of practical calculations application of the methodology for the Ukrainian financial market are provided.

Path length prediction in MANET under AODV routing: Comparative analysis of ARIMA and MLP model

Mobile Ad-hoc network (MANET) is infra-structure less collection of mobile nodes which can communicate with each other through single hop or multi-hop technique. The hop count also known as path length plays a crucial role in packet delivery, routing load, delay, etc. The path length between source destination pair nodes depends upon factors such as the mobility patterns of nodes, routing algorithm, transmission range, etc. In this article, we have tried to predict the path length between a source destination pair in MANET using Autoregressive Integrated Moving Average (ARIMA) and multilayer perceptron (MLP) models. The path length data are collected from MANETs using three different mobility models namely (i) Manhattan Grid Mobility Model (MHG), (ii) Random Way Point mobility model (RWP) and (iii) Reference Point Group Mobility Model (RPGM). This paper evaluates the predictive accuracy in forecasting the path length between source and destination nodes for Ad hoc On-Demand Distance Vector AODV routing in MANET using ARIMA model and MLP. It is found that neural networks can be effectively used in forecasting path length between mobile nodes better than statistical model and the MLP based neural network models are found to be better forecaster than ARIMA model.

Graph-Based Dependency Parsing Using Dynamic Features in Korean

In this paper, we propose a graph-based dependency parsing model using dynamic features. The dynamic features, which are derived from the partial tree during bottom-up chart parsing, play an important role in finding the correct heads. Based on the beneficial aspects of the head-final property of Korean, we suggest a variant of the CYK parsing algorithm with an O(n3) complexity which has the ability to search for the maximum spanning tree (MST) from all projective trees. Compared to Eisner's algorithm, this algorithm is time-efficient in parsing all other head-final languages. We also explain why the graph-based approaches are seldom attempted in parsing morphologically rich languages (MRLs) and suggest a morpheme-level dependency representation as a solution. Finally, we evaluate the proposed method on the KAIST corpus using two discriminative online learning methods (the averaged perceptron algorithm and passive-aggressive algorithm) and show that the trained parser achieves the best reported parsing accuracy.

Resources for Turkish morphological processing

We present a set of language resources and tools—a morphological parser, a morphological disambiguator, and a text corpus—for exploiting Turkish morphology in natural language processing applications. The morphological parser is a state-of-the-art finite-state transducer-based implementation of Turkish morphology. The disambiguator is based on the averaged perceptron algorithm and has the best accuracy reported for Turkish in the literature. The text corpus has been compiled from the web and contains about 500 million tokens. This is the largest Turkish web corpus published.

Combining Statistical and Rule-Based Approaches to Morphological Tagging of Czech Texts

Combining Statistical and Rule-Based Approaches to Morphological Tagging of Czech Texts This article is an extract of the PhD thesis (Spoustová, 2007) and it extends the article (Spoustová et al., 2007). Several hybrid disambiguation methods are described which combine the strength of hand-written disambiguation rules and statistical taggers. Three different statistical taggers (HMM, Maximum-Entropy and Averaged Perceptron) and a large set of hand-written rules are used in a tagging experiment using Prague Dependency Treebank. The results of the hybrid system are better than any other method tried for Czech tagging so far.