Recommender System Framework Research Articles

Causal Structure Learning for Recommender System

A fundamental challenge of recommender systems (RS) is understanding the causal dynamics underlying users’ decision making. Most existing literature addresses this problem by using causal structures inferred from domain knowledge. However, there are numerous phenomenons where domain knowledge is insufficient, and the causal mechanisms must be learned from the feedback data. Discovering the causal mechanism from RS feedback data is both novel and challenging, since RS itself is a source of intervention that can influence both the users’ exposure and their willingness to interact. Also for this reason, most existing solutions become inappropriate since they require data collected free from any RS. In this paper, we first formulate the underlying causal mechanism as a causal structural model and describe a general Causal Structure Learning framework for RS (CSL4RS) grounded in the real-world working mechanism. The essence of our approach is to acknowledge the unknown nature of RS intervention. We then derive the learning objective from our framework and utilize an augmented Lagrangian solver for efficient optimization. We conduct both simulation and real-world experiments to demonstrate how our approach compares favorably to existing solutions, together with the empirical analysis from sensitivity and ablation studies.

Non-Stationary Transformer Architecture: A Versatile Framework for Recommendation Systems

Recommendation systems are crucial in navigating the vast digital market. However, user data’s dynamic and non-stationary nature often hinders their efficacy. Traditional models struggle to adapt to the evolving preferences and behaviours inherent in user interaction data, posing a significant challenge for accurate prediction and personalisation. Addressing this, we propose a novel theoretical framework, the non-stationary transformer, designed to effectively capture and leverage the temporal dynamics within data. This approach enhances the traditional transformer architecture by introducing mechanisms accounting for non-stationary elements, offering a robust and adaptable solution for multi-tasking recommendation systems. Our experimental analysis, encompassing deep learning (DL) and reinforcement learning (RL) paradigms, demonstrates the framework’s superiority over benchmark models. The empirical results confirm our proposed framework’s efficacy, which provides significant performance enhancements, approximately 8% in LogLoss reduction and up to 2% increase in F1 score with other attention-related models. It also underscores its potential applicability across accumulative reward scenarios with pure reinforcement learning models. These findings advocate adopting non-stationary transformer models to tackle the complexities of today’s recommendation tasks.

Towards a Causal Decision-Making Framework for Recommender Systems

Causality is gaining more and more attention in the machine learning community and consequently also in recommender systems research. The limitations of learning offline from observed data are widely recognized, however, applying debiasing strategies like Inverse Propensity Weighting does not always solve the problem of making wrong estimates. This concept paper contributes a summary of debiasing strategies in recommender systems and the design of several toy examples demonstrating the limits of these commonly applied approaches. Therefore, we propose to map the causality frameworks of potential outcomes and structural causal models onto the recommender systems domain in order to foster future research and development. For instance, applying causal discovery strategies on offline data to learn the causal graph in order to compute counterfactuals or improve debiasing strategies.

Hybrid Recommender System for Tourism Based on Big Data and AI

With the advancement of the Internet, technology, and communication channels, the generation of tourist data has significantly increased across various sectors such as hotels, restaurants, transportation, heritage sites, tourist events, and activities. This surge is particularly notable with the rise of Online Travel Agencies (OTAs). However, the sheer volume of options provided to tourists by web search engines or specialized tourism websites often overwhelms them, burying relevant results in a sea of information "noise." This inundation hinders or slows down the decision-making process. To alleviate this issue and aid tourists in trip planning by facilitating the discovery of relevant information, numerous recommender systems have emerged. This article offers an overview of the diverse recommendation approaches employed in the tourism domain. Through this investigation, we propose an architecture and conceptual framework for a tourism recommender system founded on a hybrid recommendation approach. This system surpasses merely suggesting a list of tourist attractions tailored to individual preferences. Instead, it functions as a comprehensive trip planner, crafting detailed itineraries comprising a variety of tourism resources, customized for specific visit durations. KeyWords: Internet Development, Tourist Data Production, Trip Planning, Conceptual Framework.

Building Higher-Order Abstractions from the Components of Recommender Systems

We present a modular recommender system framework that tightly integrates yet maintains the independence of individual components, thus satisfying two of the most critical aspects of industrial applications, generality and specificity. On the one hand, we ensure that each component remains self-contained and is ready to serve in other applications beyond recommender systems. On the other hand, when these components are combined, a unified theme emerges for recommender systems. We present the details of each component in the context of recommender systems and other applications. We release each component as an open-source library, and most importantly, we release their integration under MAB2REC, an industry-strength open-source software for building bandit-based recommender systems. By bringing standalone components together, Mab2Rec realizes a powerful and scalable toolchain to build and deploy business-relevant personalization applications. Finally, we share our experience and best practices for user training, adoption, performance evaluation, deployment, and model governance within the enterprise and the broader community.

CRS-Que : A User-centric Evaluation Framework for Conversational Recommender Systems

An increasing number of recommendation systems try to enhance the overall user experience by incorporating conversational interaction. However, evaluating conversational recommender systems (CRSs) from the user’s perspective remains elusive. The GUI-based system evaluation criteria may be inadequate for their conversational counterparts. This article presents our proposed unifying framework, CRS-Que , to evaluate the user experience of CRSs. This new evaluation framework is developed based on ResQue , a popular user-centric evaluation framework for recommender systems. Additionally, it includes user experience metrics of conversation (e.g., understanding, response quality, humanness) under two dimensions of ResQue (i.e., Perceived Qualities and User Beliefs). Following the psychometric modeling method, we validate our framework by evaluating two conversational recommender systems in different scenarios: music exploration and mobile phone purchase . The results of the two studies support the validity and reliability of the constructs in our framework and reveal how conversation constructs and recommendation constructs interact and influence the overall user experience of the CRS. We believe this framework could help researchers conduct standardized user-centric research for conversational recommender systems and provide practitioners with insights into designing and evaluating a CRS from users’ perspectives.

The EU legal framework for algorithmic recommender systems: I (don’t) know it when I see it

ABSTRACT E-commerce providers increasingly rely on recommender systems to curate online consumption. EU law requirements for such recommender systems were initially introduced in privacy legislation, but recent years have seen several new provisions on this topic in consumer protection legislation and platform regulation. However, the latter regulations are limited to online intermediaries and, just like privacy legislation, they merely require businesses that use recommender systems to provide transparency on this to consumers. Transparency is not a solution for the challenges that recommender systems pose because it is difficult to achieve (especially towards consumers) and is not enough to allow users to control recommendations. Furthermore, most of these tools rely on private enforcement, which has proved to be less than effective.

Robust Privacy-Preserving Recommendation Systems Driven by Multimodal Federated Learning.

Recommendation system (RS) is an important information filtering tool in nowadays digital era. With the growing concern on privacy, deploying RSs in a federated learning (FL) manner emerges as a promising solution, which can train a high-quality model on the premise that the server does not directly access sensitive user data. Nevertheless, some malicious clients can deduce user data by analyzing the uploaded model parameters. Even worse, some Byzantine clients can also send contaminated data to the server, causing blockage or failure of model convergence. In addition, most existing researches on federated recommendation algorithms only focus on unimodality learning, ignoring the assistance of multiple modality data to promote recommendation accuracy. Therefore, this article designs an FL-based privacy-preserving multimodal RS framework. To distinguish various modality data, an attention mechanism is introduced, wherein different weight ratios are assigned to various modal features. To further strengthen the privacy, local differential privacy (LDP) and personalized FL strategies are designed to identify malicious clients and bolster the resilience against Byzantine attacks. Finally, two multimodal datasets are established to verify the effectiveness of the proposed algorithm. The superiority of our proposed techniques is confirmed by the simulation results.

Innovative Framework for Context-based Recommendation using TF-IDF and Cosine Similarity

The study investigates a novel framework for context-based recommendation systems that makes use of cosine similarity and TF-IDF (Term Frequency-Inverse Document Frequency) methods. Through the recommendation of pertinent things based on context and user preferences, recommendation systems significantly improve user experience. Recommendations from traditional systems are frequently not as good because of their inability to properly include contextual information. In order to close this gap, this work suggests a unique method that represents item and user profiles using TF-IDF and calculates cosine similarity to capture semantic similarities. The framework's ability to adjust dynamically to shifting user situations improves the accuracy of recommendations in a variety of industries, including personalized services, streaming media, and e-commerce. From a methodological point of view, the study entails using and assessing the framework on a dataset customized for context-aware recommendation situations. Comparing the approach to baseline methods, experimental results show that it is successful in enhancing recommendation quality. By incorporating strong contextual understanding, the knowledge obtained from this work advances recommendation systems and opens the door to more customized and flexible user experiences in digital contexts. The concept, methods, and main conclusions of the study are briefly described in this abstract, which also emphasizes the novel use of TF-IDF and cosine similarity for context-based recommendations.

A global user profile framework for effective recommender systems

AbstractModern Recommender Systems (RSs) compete to maintain rich user profiles that can accurately reflect user behavior, interests, and service contexts. While benefiting from an online service supported by an RS, user preferences and interests may rapidly change over time. To keep up with the changes from the user perspective, an RS should maintain the making of effective personalization as supported by robust profile construction methods. Building an effective user profile database requires exhaustive data and behavior analysis over extended periods. In this paper, we delve into traditional RS architectures to identify limitations, gaps, and opportunities for improvements in existing user profile mechanisms. To that end, a Global User Profile Framework (GUPF) is proposed towards achieving increased effectiveness. Furthermore, the adoption of the developed framework is exemplified by presenting different potential scenarios. The presented work concludes with the identification of important venues and research directions that are enabled by the proposed GUPF.

A novel evaluation framework for recommender systems in big data environments

A novel evaluation framework for recommender systems in big data environments

Analysis of using big data to achieve precise employment of college students

Abstract In order to solve the social phenomenon of difficult employment of college students, this paper constructs a college employment recommendation platform based on an improved collaborative filtering algorithm. Firstly, the framework of the employment recommendation system is designed for the user use case, system architecture, system function module, database and user portrait of the platform, respectively. The functional modules are divided into six module parts: campus recruitment management, internship management, employment information management, employment affairs management, employment analysis management, and employment guidance management. Next, the platform is analyzed to combine the recommendation algorithm of collaborative filtering algorithm and clustering technology based on users and items to realize the recommendation of employment information for college students. Then the employment platform is used to get the high stickiness of users in the employment of college students, which can retain about 62% of the number of people, and even 5% of the number of people are particularly fond of the platform. With the increase in recommendation times, the recommendation accuracy increased from 22% to 83%, and the accuracy increased by about 61%, and the comparison with the traditional employment form is much higher than the traditional method in both accuracy and recall rate by about 30%. It is concluded that using the college employment recommendation platform can realize the precise employment of college students.

Poisoning QoS-aware cloud API recommender system with generative adversarial network attack

With the proliferation and deepening of service-oriented architecture, more and more enterprises and organizations are exposing their computing functions and big data to the Internet in the form of cloud APIs to support service-oriented software development. This has resulted in a plethora of cloud APIs with similar functionality appearing on the Web, drowning users in a sea of cloud API choices. To solve this problem, quality of service (QoS)-aware recommender system is then widely applied to the selection of cloud APIs. Due to the dynamic and open network environment, the QoS-aware cloud API recommender systems are vulnerable to data poisoning attacks, where attackers inject poisoned data to skew the recommender system and make the recommendation direction follow the attacker's will. Given the lack of data poisoning attack methods and robustness analysis for existing QoS-aware cloud API recommender systems, in this work, we first built a general poisoning attack framework for QoS-aware cloud API recommender systems to elucidate and standardize the attack process. Then, we proposed a deep learning-based poison attack approach, which uses generative adversarial network (GAN) to learn the cloud API QoS data distribution of real users in an adversarial way, so as to generate high-quality fake user attack vectors. We conducted extensive experiments on real-world QoS datasets, and the experimental results show that our proposed GAN-based poisoning attack is effective and can better hide itself from being detected. In addition, we analyzed the data poisoning attack mechanism and the robustness of the cloud API recommender system based on four categories of twelve recommendation methods, thereby raising awareness about the security of cloud API recommendation and helping the recommender system defenders to develop more targeted defense strategies.

Sylvester Equation Induced Collaborative Representation Learning for Recommendation

For an actual recommendation system, it generally involves a variety of heterogeneous interactive relationships, such as the typical user-user (U2U), item-item (I2I), and user-item (U2I) interaction relationships. With the application of graph neural networks (GNNs) in embedding various interactive relations, recommendation technology has made gratifying progress in recent years, which benefits lot from its powerful ability in relation modeling. However, most of the existing GNN-based methods fail to collaboratively explore the above heterogeneous multiple interactive relationships, including the internal correlations among multiple relationships and the intrinsic association behind different relationships. As a consequence, the user's personalized preference for the items to be recommended will not be well captured. In this paper, we propose a <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</u> ylvester equation induced <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</u> ollaborative <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R</u> epresentation <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</u> earning framework (S-CRL) for recommendation system by utilizing the heterogeneous multiple interactive relationships. In particular, we ingeniously define a novel Sylvester equation to associate tactfully the multiple heterogeneous relations together. From the perspective of rating propagation, such Sylvester equation is shown theoretically to be the optimal solution of a local structure sensitive rating propagation function. Additionally, to seek more expressive embeddings about user and item, a layer-wise attention is introduced to aggregate the multi-hop information from U2U and I2I graphs, respectively, so as to promote the aggregation with the corresponding embeddings from the U2I interaction graph. Extensive experiments on three real-world datasets verify that our model achieves more favorable performance over currently representative methods.

Conceptual framework of recommendation system with hybrid method

Recommendation system relies on information of user preference and user behavior in order to recommend the useful information. The existing recommendation systems still have problems for new users and new items. This research proposes a new hybrid method to develop the conceptual framework of recommendation system that deals with new user and new movie data. The data used consists of a data from MovieLens and the internet movie database (IMDB). This work introduces a hybrid recommendation system which based on a combination of content-based filtering (CBF) and collaborative filtering (CF). Pre-filtering data is performed by finding an optimal number of clusters by calculating the total within cluster sum of square. In order to reduce the complexity of data and increase the relevance of the user-item ratings, the fuzzy c-mean (FCM) is employed. Then the similarity is calculated by using item-based method, the K-nearest neighbors and weight sum of the rating are applied. Finally, to recommend the movies, the research found that for new user data the precision is at 85% and mean absolute eror (MAE) value 2.1011. For new item data, the result of research obtains the precision at 87% and MAE value 2.0031. In conclusion, the new hybrid method developed can recommend movie efficiently.

ClayRS: An end-to-end framework for reproducible knowledge-aware recommender systems

Knowledge-aware recommender systems represent one of the most innovative research directions in the area of recommender systems, aiming at giving meaning to information expressed in natural language and obtaining a deeper comprehension of the information conveyed by textual content.Though rich and constantly evolving, the literature on knowledge-aware recommender systems is particularly scattered when considering software libraries. This makes it difficult to easily exploit advanced content representation and implement replicable experimental protocols. Accordingly, this work aims to fill in these gaps by introducing ClayRS, an end-to-end framework for replicable knowledge-aware recommender systems. ClayRS provides researchers and practitioners with the most recent state-of-the-art methodologies to build knowledge-aware content representations and also includes methods to exploit these representations in content-based recommendation algorithms. Finally, the structure of the framework also allows for building replicable pipelines to push forward the current research in the area and to develop accountable recommender systems.

Modeling Users’ Curiosity in Recommender Systems

Today’s recommender systems are criticized for recommending items that are too obvious to arouse users’ interests. Therefore the research community has advocated some ”beyond accuracy” evaluation metrics such as novelty, diversity, and serendipity with the hope of promoting information discovery and sustaining users’ interests over a long period of time. While bringing in new perspectives, most of these evaluation metrics have not considered individual users’ differences in their capacity to experience those ”beyond accuracy” items. Open-minded users may embrace a wider range of recommendations than conservative users. In this paper, we proposed to use curiosity traits to capture such individual users’ differences. We developed a model to approximate an individual’s curiosity distribution over different stimulus levels. We used an item’s surprise level to estimate the stimulus level and whether such a level is in the range of the user’s appetite for stimulus, called Comfort Zone . We then proposed a recommender system framework that considers both user preference and their Comfort Zone where the curiosity is maximally aroused. Our framework differs from a typical recommender system in that it leverages human’s Comfort Zone for stimuli to promote engagement with the system. A series of evaluation experiments have been conducted to show that our framework is able to rank higher the items with not only high ratings but also high curiosity stimulation. The recommendation list generated by our algorithm has higher potential of inspiring user curiosity compared to the state-of-the-art deep learning approaches. The personalization factor for assessing the surprise stimulus levels further helps the recommender model achieve smaller (better) inter-user similarity.

Adaptive neuro-fuzzy inference system for customizing investment type based on the potential investors’ demographics and feedback

The proposed model is an adaptive neuro-fuzzy inference recommender system that utilizes customer investment service feedback and fuzzy neural inference solutions to generate personalized investment recommendations. The model is designed to support the investment process for the customers and takes into consideration seven factors to implement the proposed investment system model through the customer or potential investor data set. These include demographic data and investment type. The model is divided into three main phases: data gathering, data analysis, and decision-making. In the data gathering phase, initial data is collected through a web-based platform, and in the data analysis phase, the potential investors' demographic criteria are extracted and grouped, and the types of investments are then clustered. The output obtained is transferred to the ANFIS layer, and investment-type recommendations are extracted for each group of potential investors. Investor feedback is received to improve and develop the system. JMP and MATLAB are used to propose the model, which serves as a framework for investment recommender systems. It demonstrates how to use this framework to offer pertinent and precise recommendations for the best sort of investment type to potential and present investors by combining the expertise of the experts and the demographic information of potential investors. Overall, this paper provides a new, novel model for investment recommender systems, which can assist investment companies, individual investors, and fund managers in their investment decisions.

Iron triangle framework for the health recommender system: perspectives from intelligent care

Iron triangle framework for the health recommender system: perspectives from intelligent care

An improved hybrid semi-stacked autoencoder for item-features of recommendation system (iHSARS)

In recent years, information overload has become a phenomenon where it makes people difficult to filter relevant information. To address issues such as high-dimensional data, cold start, and data sparsity, semi-autoencoder is one of the unsupervised deep learning methods used in the recommendation systems. It is particularly useful for reducing data dimensions, capturing latent representations, and flexibly reconstructing various parts of input data. In this article, we propose an improved hybrid semi-stacked autoencoder for item-features of recommendation system (iHSARS) framework. This method aims to show better performance of the hybrid collaborative recommendation via semi-autoencoder (HRSA) technique. Two novel elements for iHSARS’s architecture have been introduced. The first element is an increase sources of side information of the input layer, while the second element is the number of hidden layers has been expanded. To verify the improvement of the model, MovieLens-100K and MovieLens-1M datasets have been applied to the model. The comparison between the proposed model and different state-of-the-art methods has been carried using mean absolute error (MAE) and root mean square error (RMSE) metrics. The experiments demonstrate that our framework improved the efficiency of the recommendation system better than others.