Online product recommendation (OPR) systems have gained prominence in the context of e-commerce over the past years. Despite the increased research on OPR use, less attention has been paid to ...examining how decision and affective assessment of the OPR are contingent upon the product type. This study proposes and examines a recommendation-product congruity proposition based on cognitive fit and schema congruity theories. The proposition states that when the content (i.e., a stimulus-based schema) of the OPR either system-generated recommendation (SGR) or a consumer-generated recommendation (CGR) matches the brain-stored schema initiated by a particular product (either a search product or an experienced product), then a consumer would use a schema-based information assessment strategy and experience favorable decision and affective assessment of the OPR. This then affects consumers’ intentions to purchase and reuse OPR. The proposition is tested
via a
2 × 2 between-respondents factorial design of a cross-sectional survey with 482 Amazon customers. The results support the following two matching conditions of the proposition: (1) SGR describing a search product and (2) CGR explaining an experienced product, which might lead customers to perceive lower decision effort, greater decision quality, and higher enjoyment with the OPR that subsequently have a significant impact on their intentions to purchase and reuse OPR. This study expands our understanding of how recommendation-product congruence influences the consumer’s decision and affective assessment behavior and provides practical implications for the identification and presentation of the recommendation type and product type for a better customer decision.
An Internet of Things (IoT) network is prone to many ways of threatening individuals. IoT sensors are lightweight, lack complicated security protocols, and face threats to privacy and ...confidentiality. Hackers can attack the IoT network and access personal information and confidential data for blackmailing, and negatively manipulate data. This study aims to propose an IoT threat protection system (IoTTPS) to protect the IoT network from threats using an ensemble model RKSVM, comprising a random forest (RF), K nearest neighbor (KNN), and support vector machine (SVM) model. The software-defined networks (SDN)-based IoT network datasets such as KDD cup 99, NSL-KDD, and CICIDS are used for threat detection based on machine learning. The experimental phase is conducted by using a decision tree (DT), logistic regression (LR), Naive Bayes (NB), RF, SVM, gradient boosting machine (GBM), KNN, and the proposed ensemble RKSVM model. Furthermore, performance is optimized by adding a grid search hyperparameter optimization technique with K-Fold cross-validation. As well as the NSL-KDD dataset, two other datasets, KDD and CIC-IDS 2017, are used to validate the performance. Classification accuracies of 99.7%, 99.3%, 99.7%, and 97.8% are obtained for DoS, Probe, U2R, and R2L attacks using the proposed ensemble RKSVM model using grid search and cross-fold validation. Experimental results demonstrate the superior performance of the proposed model for IoT threat detection.
The use of text data with high dimensionality affects classifier performance. Therefore, efficient feature selection (FS) is necessary to reduce dimensionality. In text classification challenges, FS ...algorithms based on a ranking approach are employed to improve the classification performance. To rank terms, most feature ranking algorithms, such as the Relative Discrimination Criterion (RDC) and Improved Relative Discrimination Criterion (IRDC), use document frequency (DF) and term frequency (TF). TF accepts the actual values of a term with frequently and rarely occurring terms used in existing feature ranking algorithms. However, these algorithms focus on the number of terms in a document rather than the number of terms in the category. In this research, an alternative method to RDC, called Alternative Relative Discrimination Criterion (ARDC) was proposed, which aims to improve the accuracy and effectiveness of RDC feature ranking. Specifically, ARDC is designed to identify terms commonly occurring in the positive class. The results obtained were compared to the existing RDC methods, which are RDC and IRDC, and standard benchmarking functions such as Information Gain (IG), Pearson Correlation Coefficient (PCC), and ReliefF. The experimental results reveal that using the suggested ARDC on the Reuters21578, 20newsgroup, and TDT2 datasets provides better performance in terms of precision, recall, f-measure, and accuracy when employing well-known classifiers such as multinomial naïve Bayes (MNB), Support Vector Machine (SVM), Multilayer perceptron (MLP), k-nearest neighbor (KNN), and decision tree (DT). Another experiment was performed to validate the proposed technique, which aims to showcase the novelty of the ARDC approach. The experiment utilized the 20newsgroup dataset and employed the Relevant-Based Feature Ranking (RBFR) technique. Naïve Bayes (NB), Random Forest (RF) and Logistic Regression (LR) classifiers were used in this experiment to demonstrate the effectiveness of the suggested ARDC.
This study aims to extend expectation-confirmation model (ECM) of IS continuance based on effort-accuracy model (EAM) for predicting and explaining continuous usage of online product recommendation ...(OPR) that has been ignored in prior literature. The proposed OPR continuance model, incorporating the post-adoption beliefs of perceived usefulness, perceived decision quality and perceived decision effort, was empirically validated with data collected from an online survey of 626 existing users of the OPR. Results indicated a good explanatory power of the OPR continuance model (R2 = 62.1% of OPR continuance intention, R2 = 53% of satisfaction, R2 = 50.5% of perceived usefulness, and R2 = 9% of perceived decision effort, and R2 = 72.3% of perceived decision quality), with all major paths supported except one. We also analysed the data on the original ECM that reveals lower variances explained compared to the OPR continuance model (D6% in OPR continuance intention, D5.1% in customer satisfaction, and D3.2% in perceived usefulness). The salient effect of perceived decision quality signifies that the nature of the IS can be an important boundary condition in understanding the continuance behaviour. At a practical level, this study presents deeper insights into how to address users' satisfaction and continued patronage.
Feature selection (FS) is a critical step in many data science-based applications, especially in text classification, as it includes selecting relevant and important features from an original feature ...set. This process can improve learning accuracy, streamline learning duration, and simplify outcomes. In text classification, there are often many excessive and unrelated features that impact performance of the applied classifiers, and various techniques have been suggested to tackle this problem, categorized as traditional techniques and meta-heuristic (MH) techniques. In order to discover the optimal subset of features, FS processes require a search strategy, and MH techniques use various strategies to strike a balance between exploration and exploitation. The goal of this research article is to systematically analyze the MH techniques used for FS between 2015 and 2022, focusing on 108 primary studies from three different databases such as Scopus, Science Direct, and Google Scholar to identify the techniques used, as well as their strengths and weaknesses. The findings indicate that MH techniques are efficient and outperform traditional techniques, with the potential for further exploration of MH techniques such as Ringed Seal Search (RSS) to improve FS in several applications.
Question classification is one of the essential tasks for automatic question answering implementation in natural language processing (NLP). Recently, there have been several text-mining issues such ...as text classification, document categorization, web mining, sentiment analysis, and spam filtering that have been successfully achieved by deep learning approaches. In this study, we illustrated and investigated our work on certain deep learning approaches for question classification tasks in an extremely inflected Turkish language. In this study, we trained and tested the deep learning architectures on the questions dataset in Turkish. In addition to this, we used three main deep learning approaches (Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNN)) and we also applied two different deep learning combinations of CNN-GRU and CNN-LSTM architectures. Furthermore, we applied the Word2vec technique with both skip-gram and CBOW methods for word embedding with various vector sizes on a large corpus composed of user questions. By comparing analysis, we conducted an experiment on deep learning architectures based on test and 10-cross fold validation accuracy. Experiment results were obtained to illustrate the effectiveness of various Word2vec techniques that have a considerable impact on the accuracy rate using different deep learning approaches. We attained an accuracy of 93.7% by using these techniques on the question dataset.
PurposeThis study examines the role of continuous trust (i.e., a trust that develops over time as a result of continuous usage interactions) in determining customers' intention to continue using ...online product recommendations (OPRs).Design/methodology/approachBased on information system (IS), continuance model, and continuous trust, we propose that continuous trust will influence customers’ intention to continue OPRs’ use directly and indirectly via their satisfaction and usefulness of the OPRs. The research model is tested using data collected via an online survey from 626 existing users of OPRs in 15 different countries.FindingsThe empirical results revealed that continuous trust is shown to be a significant predictor of customers’ intention to continue OPRs use for future purchases. Additionally, the customers’ perceived confirmation and continuous trust positively influence their perceived usefulness and satisfaction with the OPRs, which subsequently influence customers’ OPRs continuous usage intention.Research limitations/implicationsThe saliency of continuous trust and usefulness of OPRs present e-retailers with potential fruitful areas to shape future usage of OPRs. In addition, e-retailers must understand that improving the OPR usefulness on its own may not lead to OPRs continuous usage until OPRs trustworthiness is not continually improved. OPRs itself may be convenient and useful, but trustworthy OPRs will pay stronger dividends for customer satisfaction and OPRs’ continuous usage.Originality/valueThe incorporation of continuous trust into the IS continuance model offers a new theoretical lens and an alternative explanation for the OPR continuous usage intention. This study stands in contrast to the large majority of research concerning initial trust and OPRs adoption, in that it focuses on continuous trust (as opposed to initial trust) and on a customers’ OPRs continuous usage intention as opposed to the initial adoption decision.
Diabetes has been an offensive condition in recent years, and it can lead to major health problems. If diabetes is not addressed, it can lead to a variety of health problems, including heart disease, ...stroke, blindness, and kidney failure. Diabetes condition must be addressed promptly to avoid a significant health risk. Machine learning algorithms can assist the doctor in identifying and diagnosing diabetes and other diseases. Different types of classifiers have been used to diagnose diabetes. To improve the performance of integrated flexible individual classifiers and lower the possibility of misclassifying a single instance, an ensemble approach named "Stacking Classifier" was developed. Several classifiers, such as Naïve Bayes, KNN, Linear regression, and decision tree (DT) were used but all these models have low accuracy. However, additional study is needed to detect diabetic condition due to a lack of major work and low accuracy. Therefore this study proposed an ensemble technique termed "Stacking Classifier" was designed to increase the performance of integrated flexible individual classifiers and reduce the probability of misclassifying a single instance. This study uses a variety of classifiers, including Naïve Bayes, KNN, Linear Discriminant Analysis, and Decision Tree, with Random Forest functioning as a Meta classifier. In terms of F-measure, Recall, Accuracy, and Precision, the proposed stacking classifier achieves a higher accuracy of 97.35 % when compared to current models such as Nave Naïve Bayes, KNN, Decision Tree, and Linear Discriminant Analysis, which are 74.60 %, 78.57 %, and 77.35 %, respectively.
Data classification is one of the most frequently used tasks carried out to label information into predefined classes. The most commonly used models for data classification are Feed Forward ...Artificial Neural Networks (FFANN), and recurrent neural networks. The connection paths of these two structures are different from each other. Generally, the trained algorithm for these structures is back propagation (BP) algorithm which has many defects. For instance, due to the uncertain number of hidden layer neuron and fixed learning rate, it is easy to fall into local minimum, and it will never reach global minimum error function, it may stay at local minimum. Therefore, to make the slow learning process faster, it is necessary to carefully select the initial weight value. Therefore Meta heuristic search techniques play an important role for selection initial weights for the network. Chicken Swarm Optimization algorithm is one of the Meta heuristic technique effectively for selecting the initial weights values to converge to the optimal solution. However, this study proposed the Chicken Swarm Optimization for efficiently learn the initial weights value of Elman Neural Network algorithm. To validate the proposed algorithm, it is compared with existing algorithms such as Back Propagation Neural Network, Artificial Bee Colony Back Propagation and Genetic Algorithm Neural Network, and verified by two classification problems namely: IRIS and 7-bit parity. Simulation results shows that proposed algorithm outperforms with existing algorithms in terms of accuracy and mean square error.
Online product recommendation (OPR) plays critical role in purchasing products, but less attention has been given to examine the differential impact of distinct recommendation sources on customer’s ...decision beliefs and behaviour. This study investigates and unfolds the distinct effects of system generated recommendation (SGR) and consumer generated recommendation (CGR) on customers’ decision effort and decision quality along with the analysis of their mediating effects on OPR reuse and purchase intentions. The study tests a conceptual model based on effort-accuracy model linking SGR and CGR to perceived decision effort and decision quality in two different product settings (search versus experience products). An online survey is conducted with Amazon customers and results of valid 482 responses reveal that users of CGR (compared to SGR) express higher decision quality, while SGR (compared to CGR) is more effective in minimizing users’ decision effort. Decision factors subsequently mediate the effects of OPR use on OPR reuse and purchase intentions. Further, CGR is found to elicit higher decision quality for experience products, while SGR is effective in minimizing decision effort for search products.