Smart Tourism using Attractive and Safe Travel Recommendation Technology

The tourism sector was among the most affected sector during the COVID-19 pandemic and has lost up to USD 5.87 billion potential revenue. Since many countries closed the borders, including Indonesia, by applying travel restrictions and thus tourists postponed their visits. Whereas vaccine distribution has shown good progress as the vaccination percentage in Jakarta and Bali has shown promising results since the majority of its population has been vaccinated, and it helps many industries, including tourism, recover. However, the pandemic might change tourist behavior. In addition, information about tourist destinations is spread poorly in various sources, and it psychologically affects tourists' decision to visit. Many works have been published to address this issue with the recommendation system. However, it does not provide geopolitical variables such as PPKM in Indonesia to ensure safeness for the tourist. Therefore, this research aims to enhance innovations in the tourism industry by considering the geopolitics factor into the system using Multiple Linear Regression. The result of this research demonstrates the effectiveness of geopolitics added variable on three different cities Jakarta, Java, and Bali. It can be implemented in a wide area in Indonesia. For further research, the proposed model can be used in a wide area in Indonesia and developed for a more comprehensive recommendation system. © 2022 IEEE.

Research and Application of Online Course Recommendation System Based on TF-IDF Algorithm

In recent years, with the rapid development of Internet technology, a large number of online learning resources have emerged. Especially affected by the COVID-19 epidemic, online learning has become a very effective learning means. However, a large number of learning platforms and massive online teaching resources have the following three problems: 1) The quality of these courses is uneven and the evaluation standards are different;2) There are so many similar courses that it is difficult for learners to distinguish them;3) These classes are lack of unity and integration, and it is hard to recommend any hierarchical, coherent and systematic course resources to learners. Therefore, a recommendation model based on TF-IDF algorithm is designed to extract personalized-featured courses, use the nearest neighbor similarity to cluster the similarity of similar courses, and conduct the featured portrait of learners to realize online courses recommendation. Combined with the model design, this paper presents a tag-based online course resource recommendation system, which can fully explore learners' explicit and implicit preferences according to course tags, and recommend satisfactory MOOC resources for them with good application value. © 2022 IEEE.

A User-POI-Guide Cost Optimization Method for Tourism Planning Considering Social Distance and User Preferences

MaaS (Mobility as a Service) itself has come into common use, and these developments have attracted keen interest from the industry in recent years. MaaS can be applied as a solution to deal with the current situation by considering the social distance. However, due to the time-share mechanism, personal assets are monopolized by specific users for a long time that cannot be shared with other users at the same time. Thus, the sharing economy companies in the tourism industry (e.g., Airbnb Experience and Huber) are in a dilemma of low productivity and high cost. In this research, we propose a new travel guide sharing service that considers the concept of social distance and user preferences. The user side only needs to select simple conditions such as travel time and the number of POIs (Point of Interest) that she/he plans to visit, meanwhile, the guide side simply inputs the POIs that she/he can guide. Furthermore, by analyzing these basic information, our proposed system can recommend the tour guides, scenic spots, and route planning to provide a real-time tour guide plan, which addressed the user's preferences and reduced the face-to-face communication to users in advance. To verify the effectiveness of our proposed method, we also ask 68 users to evaluate our system and analyze the results of questionnaires. © 2023 IEEE Computer Society. All rights reserved.

Yoga Recommendation System for the Mental Well-Being of Students using Machine Learning

Many expectations placed on students by society have made stress a part of their academic lives. Youth are susceptible to the issues brought on by academic stress since they are going through a phase of transitions in both aspects i.e personal and social. Academic stress has been shown to lower academic achievement and lower motivation toward academics. Therefore, it becomes crucial to develop appropriate and effective intervention options. In recent times, due to COVID, the utilization of online health blogs and sites recommending health, exercise, and yoga has been significantly increased. The blog will provide solution to a problem and then provide precautions to common people but they lack the dynamics to suggest yoga that can be done any person or a personalized yoga by considering their health condition and not a static article. This research work intends to develop an AI model to predict the possible practices a student can do to alleviate their problem by considering their BPM, blood pressure (both systole and diastole), sleep time and some questions related to stress. The proposed stress prediction model has achieved an accuracy of 94.4% and the yoga pose recommendation system has achieved an accuracy of 97.3%. © 2022 IEEE.

A new sentiment analysis method to detect and Analyse sentiments of Covid-19 moroccan tweets using a recommender approach.

Since the beginning of the covid-19 crisis, people from all over the world have used social media platforms to publish their opinions, sentiments, and ideas about the coronavirus epidemic and their news. Due to the nature of social networks, users share an immense amount of data every day in a freeway, which gives them the possibility to express opinions and sentiments about the coronavirus pandemic regardless of the time and the place. Moreover, The rapid number of exponential cases globally has become the apprehension of panic, fear, and anxiety among people. In this paper, we propose a new sentiment analysis approach to detect sentiments in Moroccan tweets related to covid-19 from March to October 2020. The proposed model is a recommender approach using the advantages of recommendation systems for classifying each tweet into three classes: positive, negative, or neutral. Experimental results show that our method gives good accuracy(86%) and outperforms the well-known machine learning algorithms. We find also that the sentiments of users changed from period to period, and that the evolution of the epidemiological situation in morocco affects the sentiments of users.

E-learning course recommendation based on sentiment analysis using hybrid Elman similarity

Online learning is also referred to as E-learning which has gained huge attention and attracted most people during the COVID-19 lockdowns. Due to the excess of online information, users face severe challenges and difficulties realizing the best course that is being competitive in the global market. Therefore, it is necessary to develop an online recommendation system that supports the users in selecting the finest course with E-learning. Thus, the proposed work develops a robust RS model using different approaches. Initially, the pre-processing stage is performed to reduce the presented noise in the website data. Then, the feature extraction stage is done to extract the needed features using Improved TF-IDF, W2V (Word 2 Vector), and Hybrid N-gram. Finally, Elman Minimal Redundancy Maximum Relevance and Enhanced Aquila Optimization (EMRMR_EAO) model is proposed to provide Robust course recommendations. In this work, the ERNN method is used to classify the sentiments based on the similarity measure of the MRMR model. The top course recommendation is afforded depending on the similarity scores like Jaccard similarity, cosine similarity and euclidean similarity. Also, the loss function in the classifier is reduced by optimizing the weight parameters using the EAO approach. The performance analysis shows that the proposed recommendation model obtains improved results in terms of accuracy of 99.98%, recall of 99.81%, precision of 99.65%, and F-measure of 99.95%. The comparative analysis exhibit that the proposed EMRMR_EAO model attains better performance than the other existing works in the literature. © 2022 Elsevier B.V.

Information System for Leisure Time-Management in Quarantine Conditions

The research is devoted to the study of the problem of planning leisure time during quarantine periods (forced staying at home) using information technology tools. The need for adaptation and modification of the usual forms of leisure activity to the new format has been determined. The methods of providing recommendations were studied. Using the Analytical Hierarchy Method, the optimal type of system for the implementation of the proposed solution was chosen-a recommendation system. The algorithm of the recommendation system, which offers alternatives for spending time during periods of forced staying at home, is described. A weighted hybrid mechanism was used to provide recommendations. The recommendations feature of the developed prototype of the information system is the provision of offers that contain, in addition to passive types of leisure, also active ones that take into account the characteristics of each of its users. © 2022 IEEE.

Coffee Shop Recommendation System Using an Item-Based Collaborative Filtering Approach

To inhibit the rate of transmission of the Covid-19 virus, one of the efforts made by the Indonesian government is to impose a system of limiting social activities. Thus, resulting in changes in patterns and lifestyles in a short time. Including this 'Coffee' activity. A large amount of time available due to WFH has also resulted in an increase in the number of coffee connoisseurs, including the existence of the coffee shop itself. This makes it difficult for coffee fans to choose which coffee shop is the right one to go to desire. So, a recommendation system is needed that aims to provide advice on which coffee shop to choose. The recommendation system is a system that helps users overcome overflowing information by providing specific recommendations for users and it is hoped that these recommendations can meet the wishes and needs of users. There are three types of recommendation systems based on the methods they use, namely collaborative filtering, content-based filtering, and hybrid. The method used is collaborative filtering is often used in recommendation systems. Collaborative filtering is divided into two parts, namely Item-based collaborative filtering and User-based collaborative filtering. This paper uses Item-based collaborative filtering which uses rating data between users to get recommendations. In this technique, each coffee shop that is rated by the user is checked with similar coffee shops, then combines these similar coffee shops into a list of recommendations. The test results show that the Item-based collaborative filtering method with an adjusted cosine similarity algorithm can display recommendations that are by the rating given by the customer. © 2022 IEEE.

Graph Network based Approaches for Multi-modal Movie Recommendation System

The three times increase of SonyLiv viewers during the Tokyo Olympic, the 10% hike of YouTube users during the isolation era of covid-pandemic, and the 19% growth in Netflix user count due to the fastest growth of OTT, etc. have made the digital platform's mode all-time active and specific. The hourly increase of users' interactions and the e-commerce platform's desire of letting users engage on their sites are pushing researchers to shape the virtual digital web as user specific and revenue-oriented. This paper develops a deep learning-based approach for building a movie recommendation system with three main aspects: (a) using a knowledge graph to embed text and meta information of movies, (b) using multi-modal information of movies like audio, visual frames, text summary, meta data information to generate movie/user representations without directly using rating information;this multi-modal representation can help in coping up with cold-start problem of recommendation system (c) a graph attention network based approach for developing regression system. For meta encoding, we have built knowledge graph from the meta information of the movies directly. For movie-summary embedding, we extracted nouns, verbs, and object to build a knowledge graph with head-relation-tail relationships. A deep neural network, as well as Graph attention networks, are utilized for measuring performance in terms of RMSE score. The proposed system is tested on an extended MovieLens-100K data-set having multi-modal information. Experimental results establish that only rating-based embeddings in the current setup outperform the state-of-the-art techniques but usage of multi-modal information in embedding generation performs better than its single-modal counterparts. 1. © 2022 IEEE.

The Adaptive Strategies Improving Digital Twin Using the Internet of Things

Digital twins for factories and processes are becoming more prevalent and more valuable as a result of recent technological breakthroughs and the rise of smart manufacturing. There are also more potential for closed-loop analytics with digital twins, as well as with the rise of connection, data storage, and the Industrial Internet of Things (IIoT). Some factories have employed discrete event simulations (DES) to construct digital twins that are connected to the manufacturing floor and can be monitored in real time. However, it is difficult to quantify the advantages of a digital twin that is linked to the real world. With the emergence of the new generation of mobile network (5G), Tactile Internet, as well as the deployment of Industry 4.0 and Health 4.0, multimedia systems are moving towards immersed haptic-enabled human–machine interaction systems such as the digital twin (DT). Specifically, Industry 4.0 will be using DT and robots on a large scale. This will increase human–machine and interaction to a great extent. There will be multimodal communications used to interact with digital twins and robots, especially haptics. Hence, Tactile Internet will replace the conventional Internet today. In fact, a DT system can also be extended in Health 4.0 domain to act as a COVID-19 is a COVID-19 early warning system. When a person's temperature and other symptom data are tracked in real time, it may be determined whether or not it is time to see a doctor or undergo a COVID examination. In conjunction with a COVID tracing programme, the digital twin may be able to provide further information about the virus in relation to the individual. Since there are currently no well-recognized models to evaluate the performance of these systems, to address this research lacuna, we proposed a Quality of Experience (QoE) model for DT systems con-training multi-levels of subjective, objective, and physiopsychological influencing factors. The model is itemized through a fully detailed taxonomy that deduces the perceived user’s emotional and physical states during and after consuming spatial, temporal, proximal, and ed multi-modality media between humans and machines. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Collaborative and Content-Based Filtering Hybrid Method on Tourism Recommender System to Promote Less Explored Areas

The COVID-19 pandemic has significantly impacted various areas of life, including tourism. Currently, the tourism sector is starting to recover and start its activities. However, several tourist attractions have not been explored, thus making visitors less aware of information about these tours. This affects the number of tourist visits. Therefore, there is a need of an information technology approach to promote tourism objects, including a tourist recommendation system. This study proposed a hybrid recommendation system incorporating collaborative and content based filtering. This model is proven to be able to produce good rating predictions on a recommendation system. This hybrid method uses a linear combination by calculating the rating matrix and user profile as the first step in providing rating predictions. Collaborative filtering is calculated using the cosine similarity algorithm and weighted sum algorithm, while the content-based filtering method is performed by calculating the weight of each available feature. We apply this model to the Palembang tourism dataset to the the website. This system recommends existing historical tourist attractions based on visitor criteria. The results show the existing data's effective, efficient, and accurate results. The calculation result that the rating prediction using the hybrid method is 3.203. In addition, this method can also help overcome existing cold start problems. © Roman Science Publications Inc.

An Efficient Algorithm to Recommend Covid-19 Related Solutions

With the rise of Covid-19, the open-source community has devoted a huge amount of time into developing technical solutions to stop the spread of the virus. Useful solutions like symptom trackers and extensive analysis on existing datasets are a small drop in the massive number of solutions developed by people. But with the massive number of projects or solutions, it is time consuming for a motivated person to find an appropriate solution to put his time into. Therefore, seeing the inspiring amount of work done by the open source community, we are suggesting an efficient algorithm to recommend projects that are Coronavirus related to which the user can get recommendations for projects according to their preference such as language. © 2022 IEEE.

E-learning course recommendation based on sentiment analysis using hybrid Elman similarity

Online learning is also referred to as E-learning which has gained huge attention and attracted most people during the COVID-19 lockdowns. Due to the excess of online information, users face severe challenges and difficulties realizing the best course that is being competitive in the global market. Therefore, it is necessary to develop an online recommendation system that supports the users in selecting the finest course with E-learning. Thus, the proposed work develops a robust RS model using different approaches. Initially, the pre-processing stage is performed to reduce the presented noise in the website data. Then, the feature extraction stage is done to extract the needed features using Improved TF-IDF, W2V (Word 2 Vector), and Hybrid N-gram. Finally, Elman Minimal Redundancy Maximum Relevance and Enhanced Aquila Optimization (EMRMR_EAO) model is proposed to provide Robust course recommendations. In this work, the ERNN method is used to classify the sentiments based on the similarity measure of the MRMR model. The top course recommendation is afforded depending on the similarity scores like Jaccard similarity, cosine similarity and euclidean similarity. Also, the loss function in the classifier is reduced by optimizing the weight parameters using the EAO approach. The performance analysis shows that the proposed recommendation model obtains improved results in terms of accuracy of 99.98%, recall of 99.81%, precision of 99.65%, and F-measure of 99.95%. The comparative analysis exhibit that the proposed EMRMR_EAO model attains better performance than the other existing works in the literature.

AISAR: Artificial Intelligence-Based Student Assessment and Recommendation System for E-Learning in Big Data

Educational systems have advanced with the use of electronic learning (e-learning), which is a promising solution for long-distance learners. Students who engage in e-learning can access tests and exams online, making education more flexible and accessible. This work reports on the design of an e-learning system that makes recommendations to students to improve their learning. This artificial intelligence-based student assessment and recommendation (AISAR) system consists of score estimation, clustering, performance prediction, and recommendation. In addition, the importance of student authentication is recognised in situations in which students must authenticate themselves prior to using the e-learning system using their identity, password, and personal identification number. Individual scores are determined using a recurrent neural network (RNN) based on student engagement and examination scores. Then, a density-based spatial clustering algorithm (DBSCAN) using Mahalanobis distance clustering is implemented to group students based on their obtained score values. The constructed clusters are validated by estimating purity and entropy. Student performance is predicted using a threshold-based MapReduce (TMR) procedure from the score-based cluster. When predicting student performance, students are classified into two groups: average and poor, with the former being divided into below- and above-average students and the latter into poor and very poor students. This categorisation aims to provide useful recommendations for learning. A recommendation reinforcement learning algorithm, the rule-based state–action–reward–state–action (R-SARSA) algorithm, is incorporated for evaluation. Students were required to work on their subjects according to the provided recommendations. This e-learning recommendation system achieves better performance in terms of true-positives, false-positives, true-negatives, false-negatives, precision, recall, and accuracy.

Overview of personalization approaches in MOOCs

Covid-19 has been an alarming bifurcation in the last three years. Education and learning are among the areas most affected. Online learning environments are not a choice or a model for learning modernization, but it is an obligation and a unique solution to ensure educational continuity. This has led to a growing interest in MOOCs, which reveals the importance of taking into account some appropriate information to ensure learner-centered learning to overcome the requirements of the massiveness of learners and their scattering in front of the numerous services of MOOCs. Following this direction, we propose a deep study of different approaches to personalize MOOCs. Our study aims to consider affective information as one of the main personalization parameters in the learner model, to guarantee a high-quality education with a high recommendation accuracy. © 2022 IEEE.

Investigating Students’ Performance and Motivation in Computer Programming through a Gamified Recommender System

In this study, a personalized gamified recommender system was developed to help secondary-school students in Saudi Arabia learn computer programming. This recommender system supports those students by providing personalized recommendations to address their weaknesses and increase their motivation toward computer programming. A total of 60 female secondary-school students participated in this empirical study and were divided in to an intervention and comparison group. Due to the distance learning directives imposed by the COVID-19 pandemic, the whole study was conducted online. Data were collected through a post-test to measure student performance. In addition, a learning motivation questionnaire was distributed to all the study participants to measure their motivation toward learning programming. The Instructional Materials Motivation Survey questionnaire was distributed to the experimental group to measure their level of motivation after using the recommender system. The results showed that the personalized gamified recommender system positively affected the students’ performance in the intervention group and enhanced their motivation toward learning computer programming. © 2022 Taylor & Francis Group, LLC.

CAPIRS: COVID-19-Based Application Programming Interface Recommendation System for the Developers

Context: From the past few years, Application Programming Interface (API) is widely used for mobile- and web-based application developments. Software developers can integrate third-party services into their projects to achieve their development goals efficiently using APIs;however, with the rapid increase in the number of APIs, the manual selection of Mashup-oriented API is becoming more difficult for the developer. Objective: In the COVID-19 pandemic, everyone wants an update about the latest Standard Operating Procedures (SOPs) and the latest information on COVID-19. Additionally, a software developer wants to develop an application that provides the SOPs and latest information of COVID-19;a developer can add these functionalities into an application using COVID-19-based APIs. Moreover, the current work aims at proposing a COVID-19-based API recommendation system for the developers. Method: In this study, we propose a COVID-19-based API recommendation system for developers. The recommendation system takes a developer query as input and recommends top-3 APIs and supported features, which help the developer during software development. Furthermore, the proposed COVID-19-based API recommendation system ensures the maximum participation of the developers by validating the recommended APIs and recommendation system from the expert developers using research questionnaires. Results: Additionally, the proposed COVID-19-based API recommendation system's output is validated by expert developers and evaluated on 120 expert developers' queries. In addition, experiment results show that single value decomposition achieves better prediction. Conclusion: We conclude that it is significantly important to recommend APIs along with supported features to the developer for project development, and future work is needed to take more developer's queries also to build Integrated Development Environment for the developers. © 2022 World Scientific Publishing Co.

A Survey of Learning-Based Methods for Cold-Start, Social Recommendation, and Data Sparsity in E-commerce Recommendation Systems

With the continuous development of the economy and technology, people more and more rely on online shopping, especially during the pandemic of COVID19. On the other hand, sellers display many products, so customers need to make a great effort to find suitable products to meet their needs. To reduce the efforts of customers, researchers have developed many recommendation systems for online products. In this paper, to help further study recommendation systems in e-commerce, we survey the learning-based methods for solving the cold-start problem in a recommendation, social recommendation, and data sparsity. In particular, we compare these methods' pros and cons and point out the directions for further study. © 2021 IEEE.

A Study on E-Learning and Recommendation System

Introduction: Today, technology and internet both are proliferating due to which information access is becoming easier, and is creating new challenges and opportunities in all fields, especially when working in the field of education. For example, the e-learning education system can be personalized in order to acquire knowledge level and learner’s requirements in a learning process. The learning experience, as per the individual learner’s goals, should be adopted. Background: In the current educational environment, e-learning plays a significant role. For many researchers, it has become one of the most important subjects, as through the use of e-learning, the whole education system would revolutionize. There are many areas of e-learning in which research work is being carried out, such as Mass Communication, Information and Technology (IT), Education and Distance Education. Objectives: To meet the various needs of the learners such as talents, interests and goals an e-learning system needs to be designed as a personalized learning system by considering various educational experiences. Many methods such as ontologies, clustering, classification and association rules have been used along with filtering techniques to enhance the personalization and performance of the learner. Methods: This paper presents a detailed review of the literature of previous work that has been conducted in e-learning area, especially in the recommendation system. Current research works on e-learning has been discussed in this work in order to discover the research developments in this discipline. Conclusions: One of the vital functions of the current e-learning system is creating a personalized resource recommendation system. In this paper, we reviewed some crucial papers on both e-learning and recommendation systems. Future research work of this paper would be designing efficient and precise e-learning and recommendation system to deal with the problem of substantial personalized information resources as e-learning plays a vital role in preventing virus spread during COVID-19 pandemic. © 2022 Bentham Science Publishers.

Towards the Detection of Fake News on Social Networks Contributing to the Improvement of Trust and Transparency in Recommendation Systems: Trends and Challenges

In the age of the digital revolution and the widespread usage of social networks, the modalities of information consumption and production were disrupted by the shift to instantaneous transmission. Sometimes the scoop and exclusivity are just for a few minutes. Information spreads like wildfire throughout the world, with little regard for context or critical thought, resulting in the proliferation of fake news. As a result, it is preferable to have a system that allows consumers to obtain balanced news information. Some researchers attempted to detect false and authentic news using tagged data and had some success. Online social groups propagate digital false news or fake news material in the form of shares, reshares, and repostings. This work aims to detect fake news forms dispatched on social networks to enhance the quality of trust and transparency in the social network recommendation system. It provides an overview of traditional techniques used to detect fake news and modern approaches used for multiclassification using unlabeled data. Many researchers are focusing on detecting fake news, but fewer works highlight this detection’s role in improving the quality of trust in social network recommendation systems. In this research paper, we take an improved approach to assisting users in deciding which information to read by alerting them about the degree of inaccuracy of the news items they are seeing and recommending the many types of fake news that the material represents.