Naive Bayes Learning Research Articles

Intelligent, In-Vehicle Autonomous Decision-Making Functionality for Driving Style Reconfigurations

Intelligent connected vehicles (ICVs) constitute a transformative technology attracting immense research effort and holding great promise in providing road safety, transport efficiency, driving comfort, and eco-friendly mobility. As the driving environment becomes more and more “connected”, the manner in which an ICV is driven (driving style) can dynamically vary from time to time, due to the change in several parameters associated with personal traits and with the ICV’s surroundings. This necessitates fast and effective decisions to be made for a priori identifying the most appropriate driving style for an ICV. Accordingly, the main goal of this study is to present a novel, in-vehicle autonomous decision-making functionality, which enables ICVs to dynamically, transparently, and securely utilize the best available driving style (DS). The proposed functionality takes as input several parameters related to the driver’s personal characteristics and preferences, as well as the changing driving environment. A Naive Bayes learning classifier is applied for the cognitive nature of the presented functionality. Three scenarios, with regards to drivers with different personal preferences and to driving scenes with changing environment situations, are illustrated, showcasing the effectiveness of the proposed functionality.

Classification of Non-Functional Requirements From IoT Oriented Healthcare Requirement Document.

Internet of Things (IoT) involves a set of devices that aids in achieving a smart environment. Healthcare systems, which are IoT-oriented, provide monitoring services of patients' data and help take immediate steps in an emergency. Currently, machine learning-based techniques are adopted to ensure security and other non-functional requirements in smart health care systems. However, no attention is given to classifying the non-functional requirements from requirement documents. The manual process of classifying the non-functional requirements from documents is erroneous and laborious. Missing non-functional requirements in the Requirement Engineering (RE) phase results in IoT oriented healthcare system with compromised security and performance. In this research, an experiment is performed where non-functional requirements are classified from the IoT-oriented healthcare system's requirement document. The machine learning algorithms considered for classification are Logistic Regression (LR), Support Vector Machine (SVM), Multinomial Naive Bayes (MNB), K-Nearest Neighbors (KNN), ensemble, Random Forest (RF), and hybrid KNN rule-based machine learning (ML) algorithms. The results show that our novel hybrid KNN rule-based machine learning algorithm outperforms others by showing an average classification accuracy of 75.9% in classifying non-functional requirements from IoT-oriented healthcare requirement documents. This research is not only novel in its concept of using a machine learning approach for classification of non-functional requirements from IoT-oriented healthcare system requirement documents, but it also proposes a novel hybrid KNN-rule based machine learning algorithm for classification with better accuracy. A new dataset is also created for classification purposes, comprising requirements related to IoT-oriented healthcare systems. However, since this dataset is small and consists of only 104 requirements, this might affect the generalizability of the results of this research.

CASE: A Context-Aware Security Scheme for Preserving Data Privacy in IoT-Enabled Society 5.0

This article introduces the concept of context-aware attribute learning with cipher policy-attribute-based encryption (CP-ABE) to preserve the privacy of users’ information in IoT-enabled Society 5.0. The concept of Society 5.0 pioneers an abstract system unifying different smart environments (SEs) to provide seamless services to the citizens. While serving different applications, these SEs store users’ information in the cloud engendering users’ privacy. CP-ABE is one of the conventional security systems that preserves privacy with group data accessibility. Contemporary CP-ABE solutions enforce users to manually provide their contextual information, namely, attributes, to encrypt/decrypt data. From these solutions it can be conjectured that incorrect attribute selection by a user raises the issue of unauthenticated access to information. To address these issues, we propose a scheme, named the context-aware attribute learning scheme (CASE), which autonomously learns users’ contextual information, exploiting edge intelligence, generates attributes, and reduces the post-encryption data size using the learned attributes. We examine the performance of CASE with the help of a case study on CP-ABE over smart healthcare systems (SHSs). Extensive experimental results show that CASE outperforms the existing CP-ABE-based security schemes by reducing 32%–33% average network delay, 33%–35% average energy consumption, and 31%–36% average packet loss. Additionally, we analyze the performance of attribute learning schemes using the support vector machine (SVM), decision tree (DT), and naive Bayes (NB) learning models. We observe that DT reports better performance over SVM and NB in prediction accuracy, prediction time, and clock cycles required for execution.

Classification of multi-lingual tweets, into multi-class model using Naïve Bayes and semi-supervised learning

Twitter is a social media platform which has been proven to be a great tool for insights of emotions about products, policies etc. through a 280-character message called tweet, containing direct and unfiltered emotions by a large amount of user population. Twitter has attracted the attention of many researchers owing to the fact that every tweet is by default, public in nature which is not the case with Facebook. This paper proposes a model for multi-lingual (English and Roman Urdu) classification of tweets over diversely ranged classes (non-hierarchical architecture). Previous work in tweet classification is narrowly focused either on single language or either on uniform set of classes at most (Positive, Extremely Positive, Negative and Extremely Negative). The proposed model is based on semi-supervised learning and proposed feature selection approach makes it less dependent and highly adaptive for grabbing trending terms. This makes it a strong contender of choice for streaming data. In the methodology, using Naive Bayes learning algorithm for each phase, obtained remarkable accuracy of up to 87.16% leading from both KNN and SVM models which are popular for NLP and Text classification domains.

Predicting GPA in Entrepreneurship Study Program by Using Data Mining Technique

Entrepreneurship education in universities tries to produce entrepreneurs through formal education. This relatively new study program faces many challenges in creating graduates who will become a successful entrepreneur. It is expected that with a good GPA, graduates of this study program have a higher chance of becoming successful entrepreneurs. This study tries to predict the GPA from existing datasets using data mining techniques with classification methods. There are several variables involved to predict GPA, namely, attendance, gender, school origin, motivation score, capability score, and observation score. For the last three variables, the data is taken from interview when the student went through an admission selection process. There are three algorithms tested, namely Naive Bayes, Decision Tree, and Deep Learning. Using RapidMiner, the result from model validation shows that Naive Bayes algorithm has the highest accuracy for this dataset. The model shows that although the number of female students is smaller but the number of female students whose GPA is above 3.00 is higher than male student because females are more willing to attend lectures. Despite having good capability and attitude, if the students are reluctant to attend the class then it is likely that the GPA students will be below 3.00. This study provides an overview of the use of data mining techniques to predict GPA using the Naive Bayes algorithm. In addition, it is expected that educational institutions can pay more attention to the attendance of their students in class, create teaching strategies that are able to make students present in class, especially for male students.

Resiliency Assessment in Distribution Networks Using GIS-Based Predictive Risk Analytics

A new predictive risk-based framework is proposed to increase power distribution network resiliency by improving operator understanding of the status of the grid. This paper expresses the risk assessment as the correlation between likelihood and impact. The likelihood is derived from the combination of Naive Bayes learning and Jenks natural breaks classifier. The analytics included in a geographic information system platform fuse together a massive amount of data from outage recordings and weather historical databases in just one semantic parameter known as failure probability. The financial impact is determined by a time-series-based formulation that supports spatiotemporal data from fault management events and customer interruption cost. Results offer prediction of hourly risk levels and monthly accumulated risk for each feeder section of a distribution network allowing for timely tracking of the operating condition.

A Mattress System of Recognizing Sleep Postures Based on BCG Signal

Sleep posture recognition is the core index of diagnosis and treatment of positional sleep apnea syndrome. In order to detect body postures noninvasively, we developed a portable approach for sleep posture recognition using BCG signals with their morphological difference. A type of piezo-electric polymer film sensor was applied to the mattress to acquire BCG, the discrete wavelet transform with cubic B-spline was used to extract characteristic parameters and a naive Bayes learning phase was adapted to predict body postures. Eleven healthy subjects participated in the sleep simulation experiments. The results indicate that the mean error obtained from heart rates was 0.04±1.3 beats/min (±1.96 SD). The final recognition accuracy of four basic sleep postures exceeded 97%, and the average value was 97.9%. This measuring system is comfortable and accurate, which can be streamlined for daily sleep monitoring application.

Towards perfect text classification with Wikipedia-based semantic Naïve Bayes learning

Abstract This paper suggests a novel way of dramatically improving the Naive Bayes text classifier with our semantic tensor space model for document representation. In our work, we intend to achieve a perfect text classification with the semantic Naive Bayes learning that incorporates the semantic concept features into term feature statistics; for this, the Naive Bayes learning is semantically augmented under the tensor space model where the ‘concept’ space is regarded as an independent space equated with the ‘term’ and ‘document’ spaces, and it is produced with concept-level informative Wikipedia pages associated with a given document corpus. Through extensive experiments using three popular document corpora including Reuters-21578, 20Newsgroups, and OHSUMED corpora, we prove that the proposed method not only has superiority over the recent deep learning-based classification methods but also shows nearly perfect classification performance.

Towards lifelong assistive robotics: A tight coupling between object perception and manipulation

Abstract This paper presents an artificial cognitive system tightly integrating object perception and manipulation for assistive robotics. This is necessary for assistive robots, not only to perform manipulation tasks in a reasonable amount of time and in an appropriate manner, but also to robustly adapt to new environments by handling new objects. In particular, this system includes perception capabilities that allow robots to incrementally learn object categories from the set of accumulated experiences and reason about how to perform complex tasks. To achieve these goals, it is critical to detect, track and recognize objects in the environment as well as to conceptualize experiences and learn novel object categories in an open-ended manner, based on human–robot interaction. Interaction capabilities were developed to enable human users to teach new object categories and instruct the robot to perform complex tasks. A naive Bayes learning approach with a Bag-of-Words object representation are used to acquire and refine object category models. Perceptual memory is used to store object experiences, feature dictionary and object category models. Working memory is employed to support communication purposes between the different modules of the architecture. A reactive planning approach is used to carry out complex tasks. To examine the performance of the proposed architecture, a quantitative evaluation and a qualitative analysis are carried out. Experimental results show that the proposed system is able to interact with human users, learn new object categories over time, as well as perform complex tasks.

Differentially private Naive Bayes learning over multiple data sources

For meeting diverse requirements of data analysis, the machine learning classifier has been provided as a tool to evaluate data in many applications. Due to privacy concerns of preventing disclosing sensitive information, data owners often suppress their data for an untrusted trainer to train a classifier. Some existing work proposed privacy-preserving solutions for learning algorithms, which allow a trainer to build a classifier over the data from a single owner. However, they cannot be directly used in the multi-owner setting where each owner is not totally trusted for each other. In this paper, we propose a novel privacy-preserving Naive Bayes learning scheme with multiple data sources. The proposed scheme enables a trainer to train a Naive Bayes classifier over the dataset provided jointly by different data owners, without the help of a trusted curator. The training result can achieve ϵ-differential privacy while the training will not break the privacy of each owner. We implement the prototype of the scheme and conduct corresponding experiment.

An information-theoretic filter approach for value weighted classification learning in naive Bayes

Assigning weights in features has been an important topic in some classification learning algorithms. In this paper, we propose a new paradigm of assigning weights in classification learning, called value weighting method. While the current weighting methods assign a weight to each feature, we assign a different weight to the values of each feature. The performance of naive Bayes learning with value weighting method is compared with that of some other traditional methods for a number of datasets. The experimental results show that the value weighting method could improve the performance of naive Bayes significantly.

Using differential evolution for fine tuning naïve Bayesian classifiers and its application for text classification

The Naive Bayes (NB) learning algorithm is simple and effective in many domains including text classification. However, its performance depends on the accuracy of the estimated conditional probability terms. Sometimes these terms are hard to be accurately estimated especially when the training data is scarce. This work transforms the probability estimation problem into an optimization problem, and exploits three metaheuristic approaches to solve it. These approaches are Genetic Algorithms (GA), Simulated Annealing (SA), and Differential Evolution (DE). We also propose a novel DE algorithm that uses multi-parent mutation and crossover operations (MPDE) and three different methods to select the final solution. We create an initial population by manipulating the solution generated by a method used for fine tuning the NB. We evaluate the proposed methods by using their resulted solutions to build NB classifiers and compare their results with the results of obtained from classical NB and Fine-Tuning Naïve Bayesian (FTNB) algorithm, using 53 UCI benchmark data sets. We name these obtained classifiers NBGA, NBSA, NBDE, and NB-MPDE respectively. We also evaluate the performance NB-MPDE for text-classification using 18 text-classification data sets, and compare its results with the results of obtained from FTNB, BNB, and MNB. The experimental results show that using DE in general and the proposed MPDE algorithm in particular are more convenient for fine-tuning NB than all other methods, including the other two metaheuristic methods (GA, and SA). They also indicate that NB-MPDE achieves superiority over classical NB, FTNB, NBDE, NBGA, NBSA, MNB, and BNB.

Attribute weighting for averaged one-dependence estimators

Averaged one-dependence estimators (AODE) is a type of supervised learning algorithm that relaxes the conditional independence assumption that governs standard naive Bayes learning algorithms. AODE has demonstrated reasonable improvement in terms of classification performance when compared with a naive Bayes learner. However, AODE does not consider the relationships between the super-parent attribute and other normal attributes. In this paper, we propose a novel method based on AODE that weighs the relationship between the attributes called weighted AODE (WAODE), which is an attribute weighting method that uses the conditional mutual information metric to rank the relations among the attributes. We have conducted experiments on University of California, Irvine (UCI) benchmark datasets and compared accuracies between AODE and our proposed learner. The experimental results in our paper show that WAODE exhibits higher accuracy performance than the original AODE.

Effective multi-objective naïve Bayes learning for cross-project defect prediction

Software defect prediction predicts fault-prone modules which will be tested thoroughly. Thereby, limited quality control resources can be allocated effectively on them. Without sufficient local data, defects can be predicted via cross-project defect prediction (CPDP) utilizing data from other projects to build a classifier. Software defect datasets have the class imbalance problem, indicating the defect class has much fewer instances than the non-defect class does. Unless defect instances are predicted correctly, software quality could be degraded. In this context, a classifier requires to provide high accuracy of the defect class without severely worsening the accuracy of the non-defect class. This class imbalance principle seamlessly connects to the purpose of the multi-objective (MO) optimization in that MO predictive models aim at balancing many of the competing objectives. In this paper, we target to identify effective multi-objective learning techniques under cross-project (CP) environments. Three objectives are devised considering the class imbalance context. The first objective is to maximize the probability of detection (PD). The second objective is to minimize the probability of false alarm (PF). The third objective is to maximize the overall performance (e.g., Balance). We propose novel MO naive Bayes learning techniques modeled by a Harmony Search meta-heuristic algorithm. Our approaches are compared with single-objective models, other existing MO models and within-project defect prediction models. The experimental results show that the proposed approaches are promising. As a result, they can be effectively applied to satisfy various prediction needs under CP settings.

Experimental analysis of naïve Bayes classifier based on an attribute weighting framework with smooth kernel density estimations

Naive Bayes learners are widely used, efficient, and effective supervised learning methods for labeled datasets in noisy environments. It has been shown that naive Bayes learners produce reasonable performance compared with other machine learning algorithms. However, the conditional independence assumption of naive Bayes learning imposes restrictions on the handling of real-world data. To relax the independence assumption, we propose a smooth kernel to augment weights for the likelihood estimation. We then select an attribute weighting method that uses the mutual information metric to cooperate with the proposed framework. A series of experiments are conducted on 17 UCI benchmark datasets to compare the accuracy of the proposed learner against that of other methods that employ a relaxed conditional independence assumption. The results demonstrate the effectiveness and efficiency of our proposed learning algorithm. The overall results also indicate the superiority of attribute-weighting methods over those that attempt to determine the structure of the network.

A gradient approach for value weighted classification learning in naive Bayes

Feature weighting has been an important topic in classification learning algorithms. In this paper, we propose a new paradigm of assigning weights in classification learning, called value weighting method. While the current weighting methods assign a weight to each feature, we assign a different weight to the values of each feature. The proposed method is implemented in the context of naive Bayesian learning, and optimal weights of feature values are calculated using a gradient approach. The performance of naive Bayes learning with value weighting method is compared with that of other state-of-the-art methods for a number of datasets. The experimental results show that the value weighting method could improve the performance of naive Bayes significantly.

Assessment of Seismic Building Vulnerability from Space

This paper quantitatively evaluates the suitability of multi-sensor remote sensing to assess the seismic vulnerability of buildings for the example city of Padang, Indonesia. Features are derived from remote sensing data to characterize the urban environment and are subsequently combined with in situ observations. Machine learning approaches are deployed in a sequential way to identify meaningful sets of features that are suitable to predict seismic vulnerability levels of buildings. When assessing the vulnerability level according to a scoring method, the overall mean absolute percentage error is 10.6%, if using a supervised support vector regression approach. When predicting EMS-98 classes, the results show an overall accuracy of 65.4% and a kappa statistic of 0.36, if using a naive Bayes learning scheme. This study shows potential for a rapid screening assessment of large areas that should be explored further in the future.

Joint decision and Naive Bayes learning for detection of space multi-target

In the photoelectric tracking system, the detection of space multi-target is crucial for target localization and tracking. The difficulties include the interferences from CCD smear and strong noise, the few characteristics of spot-like targets and the challenge of multiple targets. In this paper, we propose a hybrid algorithm of joint decision and Naive Bayes (JD-NB) learning, and present the duty ratio feature to discriminate the target and smear blocks. Firstly, we extract the proper features and train the parameters of the Naive Bayes classifier. Secondly, target blocks are preliminarily estimated with the Naive Bayes. Lastly, the 4-adjacent blocks of the candidate target blocks are jointed to analyze the distribution pattern and the true target blocks are secondarily extracted by the method of pattern matching. Experimental results indicate that the proposed JD-NB algorithm not only possesses a high recognition rate of better than 90% for the target block, but also effectively overcomes the disturbance of the smear block. Moreover, it performs well in the detection of small and faint targets when the SNR of the block is higher than about 0.014.

Cell-based intrusion detection using wireless mesh network

In order to obtain optimal surveillance detection we may need to deploy large number of Wi-Fi sensors and cameras which relatively incur huge cost. Fortunately, wireless network infrastructure has potential to enhance the detection without need to massively deploy extra physical sensors by exploiting the triangulation of Wi-Fi signal interference. This paper presents experimental results that demonstrate the accuracy of Wi-Fi signal strength mapping approach to detect an existence of a human through a prediction model which is generated by Naive Bayes learning algorithm. This research introduces the cell-based intrusion detection offer by the advantages of the wireless mesh network over the traditional wireless network infrastructure. The result shows that the prediction accuracy of whether a subject is exist inside or outside of the wireless mesh imaginary boundary is at 90%. Therefore the proposed cell-based intrusion detection using wireless mesh network has a great potential to be used to enhance surveillance system detection without need to massively deploy extra physical sensors. Henceforth, our proposed cell-based intrusion detection through a wireless mesh network has a great potential to enhance current home surveillance system deployment in a large residential area.

A K-Means and Naive Bayes Learning Approach for Better Intrusion Detection

Intrusion Detection Systems (IDS) have become an important building block of any sound defense network infrastructure. Malicious attacks have brought more adverse impacts on the networks than before, increasing the need for an effective approach to detect and identify such attacks more effectively. In this study two learning approaches, K-Means Clustering and Naive Bayes classifier (KMNB) are used to perform intrusion detection. K-Means is used to identify groups of samples that behave similarly and dissimilarly such as malicious and non-malicious activity in the first stage while Naive Bayes is used in the second stage to classify all data into correct class category. Experiments were performed with KDD Cup '99 data sets. The experimental results show that KMNB significantly improved and increased the accuracy, detection rate and false alarm of single Naive Bayes classifier up to 99.6, 99.8 and 0.5%.