Recommender Research Articles

Behavioral recommendation engine driven by only non-identifiable user data

Most recommendation systems utilize personal data to device personalized recommendations for users. Even though it seems favorable, security risks like data breaches are inevitable. This research proposes a novel reinforcement learning ‘approach’ to recommend users without collecting identifiable data. With only user activity on a session, our proposed method can model and track user behavior and formulate a recommendation system. We conclude that our algorithms demonstrate positive results in capturing user behavior without collecting private data of any kind from the user. The research is two folds. On one hand, we experiment using traditional reinforcement learning techniques (MDP, Q-learning), and on the other hand, we use deep reinforcement learning algorithms (DQN, DDQN, and D3QN) on a movie recommendation scenario. Interestingly, we observe that MDP and D3QN works comparatively better on movie recommendations.

Collaborative Learning based Recommendation System for Content Streaming Platform using Non-Negative Matrix Factorization Clustering

The Recommender System helps users find relevant content by leveraging their interest models. This paper focuses on content-based recommender systems, which determine recommendations based on previously selected products without needing user evaluations. These systems can now use machine learning to profile both consumers and items. We introduce a collaborative learning method for content streaming platforms using Non-Negative Matrix Factorization Clustering (NNMFC). The research employs a sliding window technique for clustering, implemented with gradient descent and further optimized using Particle Swarm Optimization (PSO). The model's performance was examined through its Root Mean Square Error (RMSE) across three techniques: gradient descent, sliding window gradient descent, and sliding window PSO. The result analysis of the proposed work achieved the lowest RMSE of 0.90, outperforming existing state-of-the-art approaches such as Collaborative Filtering, K-Mean, and UPCSim. The Sliding Window PSO emerged as the most proficient, presenting a notable 9% enhancement over existing methods.

The application of business analytics in personalized customer experience

Purpose: The purpose of this publication is to present the applications of usage of business analytics in personalized customer experience. Design/methodology/approach: Critical literature analysis. Analysis of international literature from main databases and polish literature and legal acts connecting with researched topic. Findings: The applications of business analytics in personalized customer experiences span data collection, segmentation, predictive analytics, and multi-channel personalization, delivering greater customer satisfaction, revenue, and competitive advantage. Achieving this transformation requires the use of various software solutions, including CRM systems, marketing automation tools, analytics platforms, predictive analytics software, and recommendation engines. Harnessing the power of data and analytics is the key to gaining a competitive edge in this customer-centric era. Businesses that tailor their offerings to each customer's unique preferences and needs can not only meet but exceed customer expectations, ensuring growth and long-term success. Ultimately, personalized customer experiences, driven by business analytics, are not merely a response to customer demands; they are the path to unlocking the full potential of customer relationships and establishing a strong position in the ever-evolving marketplace. Keywords: business analytics, Industry 4.0, digitalization, artificial intelligence, real-time monitoring; customization, personalization. Category of the paper: literature review.

Hybrid Approaches for Stocks Prediction and Recommendation System

Hybrid approaches to stock prediction and recommendation are a critical area of research for individual investors and financial institutions. Traditional methods have limitations, leading to the emergence of hybrid models. This paper reviews current research on hybrid models, including GAN-based, LSTM-based, and neural network-based models, Soft Computing based, GRU based models to provide optimal results, for stock recommendation techniques include sentiment analysis, which uses natural language processing to analyze news articles and social media posts, and network analysis, which examines the relationships between stocks to identify stocks likely to move together. It also discusses evaluation metrics used to assess the performance of these models and then it provides the generalize pipelines that can be kept in mind while researching and developing a recommender engine, it also shows the future direction in order to build the hybrid recommenders as well as predictors, making it a valuable contribution to the stock prediction and recommendation field.

MOVIE RECOMMENDATION SYSTEM: A REVIEW

Whether we are choosing a series on an OTT platform or searching for a movie, online recommendation engines have influenced our decisions. They are far from perfect and are still very much in the early phases of development. Specifically, we provide bandy movie recommendation algorithms in this study. Furthermore, we discuss key exploratory articles that have helped these systems overcome certain challenges and critically evaluate some work on movie recommendation systems. Even though there have been improvements, recommendation systems still require effort to improve their ability to provide reliable suggestions for a larger variety of operations.

Product Personalized Recommendation Engine Based on Collaborative Filtering Algorithm

Nowadays, information plus Internet technology is continuously developing, and people have entered the era of information overload from the era of lack of information access. In this context, it is difficult to find one's own "tailor-made" information from the huge amount of information. In this context, the recommendation system came into being. This article uses the order data under an e-commerce platform, uses user-based and item-based collaborative filtering algorithms to personalize products, and builds a product recommendation system. Through the feasibility of actual application scenarios and the prediction score of the model, a series of evaluations were made on the algorithm, and reliable and effective prediction results were obtained from the three indicators of accuracy, recall and F1 score. The final recommendation effect of the TopN recommendation list obtained was significant, and the recommendation efficiency could reach 85%.Provide some ideas for the research and development direction of modern recommendation algorithms.

Toward a Superintelligent Action Recommender for Network Operation Centers Using Reinforcement Learning

Today’s Network Operation Centres (NOC) consist of teams of network professionals responsible for monitoring and taking actions for their network’s health. Most of these NOC actions are relatively complex and executed manually; only the simplest tasks can be automated with rules-based software. But today’s networks are getting larger and more complex. Therefore, deciding what action to take in the face of non-trivial problems has essentially become an art that depends on collective human intelligence of NOC technicians, specialized support teams organized by technology domains, and vendors’ technical support. This model is getting increasingly expensive and inefficient, and the automation of all or at least some NOC tasks is now considered a desirable step towards autonomous and self-healing networks. In this article, we investigate whether an autonomous NOC can achieve superintelligence; i.e., it can recommend or take actions that lead to better results than those achieved by the collective human intelligence, in this case rules designed by human experts. Our investigation is inspired by the superintelligence achieved in computer games recently. Specifically, we build an Action Recommendation Engine based on Reinforcement Learning, a type of Artificial Intelligence method, and we train it with expert rules and lets it explore actions by itself. We then show that it can learn new and more efficient strategies that outperform expert rules designed by humans. This can be used in face of network problems to either quickly recommend actions to NOC technicians or autonomously take actions for fast recovery.

Adaptive Fusion and Transfer Learning for Enhanced E –Commerce Recommendations

Adaptive Fusion and Transfer Learning for Enhanced E-commerce Recommendations (AFETLER) is a holistic combination of some state-of-the-art recommendation frameworks. Most of the existing frameworks, be it Matrix Factorisation model, Deep learning-based model, Factorisation Machines and Two tower recommendation models, used in real world recommendation engines lack diversity and serendipity. Also, cold start problem is one that downgrade the recommendation algorithm. Defining the optimisation goal and closed feedback improvement loop is also something that is not given primary weightage.To overcome the above stated challenges and provide a novel approach to overcome those, we propose the AFETLER model. AFETLER incorporates an adaptive fusion technique that dynamically combines recommendations from several sources such as new arrivals forecast, Click-Through Rate (CTR) prediction, and popularity estimation. This fusion technique improves accuracy by giving varying weights to each source based on the user's context, history preferences, and current behaviour, resulting in highly relevant and tailored buying recommendations. To address the cold-start problem and improve recommendations for new items, AFETLER employs a transfer learning mechanism inspired by the Adversarial Two-tower Neural Network (ATNN) framework, utilizing knowledge from existing items' interactions to extract valuable features from new items' profiles. Furthermore, AFETLER leverages hierarchical attention methods to effectively capture user-item interaction patterns, boosting feature extraction and overall recommendation quality through dynamic weighing of interaction relevance.Proposed AFETLER model outperforms some of the existing recommendation models by an average 2%-5% on metrics such a precision, ROC-AUC, and Top-K recommendation accuracy. Further, we also see point reduction in Mean absolute error and mean squared error.

Edge detection and graph neural networks to classify mammograms: A case study with a dataset from Vietnamese patients

Mammograms are breast X-ray images and they are used by doctors, among other purposes, as an effective means of detecting breast cancer. Screening mammography is crucial since it allows doctors to understand better the situation and have suitable intervention. The classification of medical modalities is a prerequisite for development of computer-aided diagnosis tools in healthcare, and various techniques have been proposed to automatically classify from mammography images. Though there have been several tools developed, they have been mostly validated with data collected from Western women. Based on our initial investigations, breast anatomy in Vietnamese women differs from that of Western women, due to denser breast tissue. In this paper, we propose MammoGNN – a practical solution to the classification of mammograms using the synergy between image processing techniques and graph neural networks. First, a well-founded edge detection algorithm was applied to provide input for the recommendation engine. Afterward, we empirically experimented to select suitable graph neural networks to manage the training and prediction. A mammogram dataset was curated from 2,351 Vietnamese women to validate the conceived tool. By several testing instances, MammoGNN obtains a maximum accuracy of 100%, precision and recall of 1.0 on independent and shuffle test sets for both classification of BI-RADS scores and breast density types. The experimental results also demonstrate that our proposed approach obtains an optimal prediction performance on the considered datasets, outperforming different baselines. We anticipate that the proposed approach can be deployed as a non-invasive pre-screening tool to assist doctors in performing their diagnosis activities.

Personalized Recommendation of Educational Resource Information Based on Adaptive Genetic Algorithm

The abundance of online educational resources has made it increasingly difficult for students to identify the correct learning materials in recent years. Overcoming the information overload that has emerged in the new education systems is possible via a tailored recommendation system. It encourages students to look for new ways to get around the subject matter and to use information from all across the world. Because of this, many academics are working to create learning systems that incorporate methods for creating a unique learning experience for each user. Therefore, our proposed approach was to create an appropriate learning route for each student, and they are using Educational Resource Information Based on an Adaptable Genetic Algorithm(ERI-AGA). Evidence from studies shows that the suggested technique can provide relevant course materials for students based on the specific needs of students to help them study better in a Web-based system. Personal recommendation engine, pre-processing and learning-based model development, and implementation of the recommendation system will be researched. Participatory budgeting PB-level data storage and processing as well as the ability to suggest in real time will be studied. The capacity to make real-time suggestions and the storing and processing of PB-level data will be investigated. It was critical to check the system’s availability by running associated tasks and performance tests. The comparison values demonstrated that ERI-AGA was a reliable and accurate assessment procedure.

Collaborative And Popularity Based Book Recommender System

Utilization of online websites to shop for a range of products has been frequent in our day to day lives. As a result, consumer demand is becoming more diverse, making it difficult for a general store to deliver the proper products based on the tastes of its customers. To deliver a favorable buying experience for the consumer, these E-commerce websites use various recommendation system algorithms. Recommendation systems are a tool for dealing with this problem; they allow you to meet consumer’s demands and expectations while also attracting new ones. A product recommendation system is essentially a filtering system that suggests particular things to customers depending on their interests. Recommendation systems have exploded in popularity in recent years with applications in music, news, movies search queries and others. The bulk of today’s E- commerce sites such as Amazon, Flipkart ,Myntra, make use of their own recommendation algorithms to better offer their customers with products they are likely to like .Recommendation engines are data filtering technologies that use algorithms and data to suggest the most relevant items to the user.

Rendering automatic bokeh recommendation engine for photos using deep learning algorithm

Abstract Automatic bokeh is one of the smartphone’s essential photography effects. This effect enhances the quality of the image where the subject background gets out of focus by providing a soft (i.e., diverse) background. Most smartphones have a single rear camera that is lacking to provide which effects need to be applied to which kind of images. To do the same, smartphones depend on different software to generate the bokeh effect on images. Blur, Color-point, Zoom, Spin, Big Bokeh, Color Picker, Low-key, High-Key, and Silhouette are the popular bokeh effects. With this wide range of bokeh types available, it is difficult for the user to choose a suitable effect for their images. Deep Learning (DL) models (i.e., MobileNetV2, InceptionV3, and VGG16) are used in this work to recommend high-quality bokeh effects for images. Four thousand five hundred images are collected from online resources such as Google images, Unsplash, and Kaggle to examine the model performance. 85% accuracy has been achieved for recommending different bokeh effects using the proposed model MobileNetV2, which exceeds many of the existing models.

Application of Internet of Things technology in Human resource recommendation engine

In order to avoid the phenomenon of information overload in the field of human resources, the author proposes a human resource recommendation engine based on Internet of Things technology. Firstly, the application fields of the Internet of Things are introduced, and the human resource recommendation engine is designed, use a mixed referral approach in the HR field. We chose a hybrid recommendation strategy of PLSA and content based on the Internet of Things. The author introduces the human resource recommendation and content-based human resource recommendation of PLSA based on the Internet of Things respectively, on the basis of the above introduction, through the analysis of the existing hybrid recommendation strategy, a weighted hybrid method of PLSA based on the Internet of Things and content-based recommendation algorithm is proposed. In order to verify the feasibility of the proposed IoT-based PLSA and content-based HR recommendation method, we conducted a multi-step experiment. We selected 50 graduating students from A University to participate in this experiment, and the results show that in order to verify the overall effect of the algorithm, with accuracy and recall as evaluation criteria, we conducted experiments on recommendation algorithms of PLSA based on Internet of Things, content-based recommendation algorithms, and recommendation algorithms based on PLSA and content weighting mixture, where, the number of z in PLSA algorithm is 10, the value of β in mixed recommendation is 0.6, the size of N in TopN recommendation is N=5,10,20, and 30. Experiments were carried out on the accuracy and recall of the three recommendation algorithms, the number of z in PLSA algorithm is 10, the value of β is 0.6, and the size of N in TopN recommendation is N=3,5,7,9. The recommendation effect of PLSA based on the Internet of Things is better than that of content-based recommendation algorithm, and the recommendation effect of the combination of the two algorithms is better than either of them.

The personalized recommendation for OTA flight cancellation and change services during the pandemic

The COVID-19 pandemic has had a dramatic impact on people’s travels. Due to the recurrent pandemic and regionally different policies in China, travelers must pay a lot for flight cancellations and changes. To accommodate this, online travel agencies (OTA) can provide a more flexible ancillary as a supplement to the airline company's services. Here, we introduced the upgraded all-in-one (AIO) service package, which offers compensation for flight delays, changes, or refund. We also designed a dynamic recommendation engine (DRE), which can make real-time personalized recommendations. Backed by AB testing, the machine learning-based DRE not only raises the package attach rate without interrupting the flight ordering process, but also helps the customers cut cost when making flight cancellations or changes.

Open-Source Software in the Classroom: Empowering Students to Self-Select Projects to Contribute

Contribution: Instructors are leveraging open-source software (OSS) as a way to experience authentic examples of software problems with their students. Recommender engines might assist students in selecting the right project based on metrics mined from project repositories (e.g., GitHub). This vision is realized through GitMate, a GitHub-based recommender for supporting students in their OSS selection. Background: Contributing to OSS is a valuable way to immerse students into the realities of software development. When it comes to OSS selection, self-selection seems to be the most engaging alternative. Yet, students lack the time (and skills) to analyze project facets and draw comparisons among OSS projects. Research Questions: How can students be assisted to select a good OSS project to contribute to? Specifically, how would a recommender system might help? The envisioned intervention should be useful not only in finding the right project but also challenging students’ initial selections with other alternatives, spurring reflection. Methodology: The aim is to act upon a dependent variable (mind changing in project selection) through an independent variable (project comparison). This is achieved through GitMate, a recommender system on top of GitHub. Its search facilities are used for students to locate three projects at their wish. Next, GitMate recommends similar projects based on the project facets (e.g., number of committers, commits, and stars), mined from GitHub. Pondering the importance of distinct facets, students can now tradeoff different projects. The experiment checks whether students change their first choice. Findings: The results indicate that GitMate helps students compare GitHub projects to the extent of making them change their first choice. Nearly, 80% of the students changed at least one project as a result of using GitMate. This seems to suggest GitMate being effective on its goal: facet-based comparison thinking during OSS selection.

Mapping Growth of the Russian Domestic Propaganda Apparatus on Telegram

As a part of Western sanctions against Russia, big social media companies i.e., Facebook, Twitter, and YouTube, have been limiting their functionalities available to Russians and suspending and limiting reach of Russian influencers. Many Russian propagandist, aware of the risk of becoming deplatformed, were already looking for social media platforms with less moderation before the February 24th. With many Russians using Telegram as their preferred communicator, Telegram became the natural choice. Russian government, by blocking access to Western social media platforms by Russian users and by directing state influencers to migrate their domestic audiences to Telegram, has accelerated this trend. Despite many advantages, Telegram is limited in ease with which new content and channels can be discovered. Telegram search for channels is based on exact string matching and Telegram has no content search or content recommendation functionality built into it. Telegram’s advertising mechanism is still in its infancy. To enable quick growth of the key state propaganda channels, Russians employed several tactics. They have selected a few existing influencers, such as Vladimir Soloviev, to substitute for the absent search and recommendation engine. Soloviev’s Telegram channels has been heavily advertised on legacy media, starting with TV stations carrying his programs, to serve as an onramp to Telegram. Soloviev has regularly featured reviews of other smaller channels, endorsing, and promoting their pro-conflict content, becoming the missing recommendation engine for Russians. All the above resulted in creating a vast, Telegram-based apparatus of domestic state propaganda. In this paper, we trace the growth of Russian pro-conflict Telegram communities and quantify the impact of centralized endorsement by state influencers on the growth of other smaller, specialized Telegram channels.

Enhancing Ecommerce Recommenders with Dual Transformer Models

The growing complexity and scale of ecommerce platforms have created a need for more advanced and precise recommendation systems. Traditional recommendation approaches, such as collaborative filtering and content-based models, while effective, often struggle to fully capture the evolving preferences and behaviours of users in dynamic online marketplaces. To address these limitations, this paper explores the application of dual transformer models in enhancing ecommerce recommender systems. Transformer models, originally developed for natural language processing, offer powerful attention mechanisms that allow for the processing of sequential and contextual data, providing a more nuanced understanding of user behaviour. This study introduces dual transformer models, which simultaneously process both user interaction history and item features to improve recommendation accuracy. By capturing the intricate relationships between users and items from both dimensions, dual transformer models offer a bidirectional approach that can adapt to changing user preferences in real time. This enables ecommerce platforms to deliver more relevant, personalized, and context-aware product recommendations. The use of dual transformers in recommendation systems presents several benefits, including improved scalability, enhanced personalization, and increased user engagement. By addressing the shortcomings of traditional recommender systems, dual transformer models have the potential to revolutionize the ecommerce industry, leading to better user satisfaction and higher conversion rates. This paper highlights the potential of these models in improving the overall effectiveness of ecommerce recommendation engines, making them a valuable tool for modern online shopping platforms.

IoT-based Recommender Engine for Yielding Better Crops

Background: Artificial Intelligence (AI) operations have evolved over the last two decades in improvising the operations for an agriculture-based economy. The operations in this area face many challenges in maximizing their crop yield while facing the challenges like deficient soil treatment, problems from pests and crop-based diseases. Technical challenges for the requirements of real-time data which can be further transformed to big data. Its impact is fetching low yield, because of the knowledge gap between farmers and technology. These points are the key motivators of introducing an ecosystem with artificial intelligence to agriculture in this research work. IoT devices are capable to generate large amounts of data that could be transformed into information about environmental parameters like the temperature of the field, which acts as an engine to provide data. All the data shall be collected, stored and further analyzed for better decision-making. One way the business uses the data collected is by nourishing it into the AI systems, which grasp the IoT data and further use it to make predictions. Methods: This research work enables to cater the solution for supporting the farmers with IT as an enabler using data analytics on the collected information. It uses a web application that would help to monitor the soil fertility and suggest the producers to select the best crop(s) that can be grown in that geographical region. Result: The result from our crop recommender engine successfully predicted which crop(s) we could choose to grow in our gardens or farm fields. An extensive soil study along with meteorological models helped us to launch a smart agricultural system in a much smarter way. This has further helped to bridge the gap between production and quantity yield.

Movie Recommender System using Machine Learning

In this research, we propose a movie recommender system that can recommend movies to both new and existing customers. It searches movie databases for all of the relevant data, such as popularity and beauty that is required for a recommendation. We apply both content-based and collaborative filtering and evaluate their advantages and disadvantages. To build a system that delivers more exact movie recommendations, we employ hybrid filtering, which is a combination of the outcomes of these two processes. The recommendation engines are also used for business purposes and to make strategies for organizations. Due to the growing demands of customers and user’s recommendation systems plays a huge role. These recommender systems also help us to utilize our time in the busy world by giving us more relevant searches. These systems are generally used with the movie’s websites or with many commercial applications and are of great use. This type of recommendation system can be also used for precise results. It will make movies suggestions more relevant as per the need of the users.

Performance appraisal of Trichoderma viride based novel tablet and powder formulations for management of Fusarium wilt disease in chickpea.

In developing a Trichoderma viride-based biocontrol program for Fusarium wilt disease in chickpea, the choice of the quality formulation is imperative. In the present study, two types of formulations i.e. powder for seed treatment (TvP) and tablet for direct application (TvT), employing T. viride as the biocontrol agent, were evaluated for their ability to control chickpea wilt under field conditions at three dosages i.e. recommended (RD), double of recommended (DD) and half of recommended (1/2 RD). A screening study for the antagonistic fungi strains based on volatile and non-volatile bioassays revealed that T. viride ITCC 7764 has the most potential among the five strains tested (ITCC 6889, ITCC 7204, ITCC 7764, ITCC 7847, ITCC 8276), which was then used to develop the TvP and TvT formulations. Gas Chromatography-Mass Spectrometry (GC-MS) analysis of volatile organic compounds (VOCs) of T. viride strain confirmed the highest abundance of compositions comprising octan-3-one (13.92%), 3-octanol (10.57%), and 1-octen-3-ol (9.40%) in the most potential T. viride 7764. Further Physico-chemical characterization by standard Collaborative International Pesticides Analytical Council (CIPAC) methods revealed the optimized TvP formulation to be free flowing at pH 6.50, with a density of 0.732 g cm-3. The TvT formulation showed a pH value of 7.16 and density of 0.0017 g cm-3 for a complete disintegration time of 22.5 min. The biocontrol potential of TvP formulation was found to be superior to that of TvT formulation in terms of both seed germination and wilt incidence in chickpea under field conditions. However, both the developed formulations (TvP and TvT) expressed greater bioefficacy compared to the synthetic fungicide (Carbendazim 50% WP) and the conventional talc-based formulation. Further research should be carried out on the compatibility of the developed products with other agrochemicals of synthetic or natural origin to develop an integrated disease management (IDM) schedule in chickpea.