Learning User Profiles Research Articles

Incremental learning user profile and deep reinforcement learning for managing building energy in heating water

Deep reinforcement learning (DRL) has garnered growing attention as a data-driven control technique in the field of built environments. However, the existing DRL approaches for managing water systems cannot consider information from multiple time steps, are prone to overestimation, fall into the problem of locally optimal solutions, and fail to cope with time-varying environments, resulting in an inability to minimize energy consumption while considering water comfort and hygiene of occupants. Therefore, this study proposes an incremental learning user profile and deep reinforcement learning (ILUPDRL) method for controlling hot water systems. This study employs hot water user profiles to reflect the hot water demand (HWD) habits. The proposed ILUPDRL addresses the challenges arising from evolving HWD through incremental learning of hot water user profiles. Moreover, to enable the ILUPDRL to consider information from multiple time steps, this study proposes the recurrent proximal policy optimization (RPPO) algorithm and integrates the RPPO into the ILUPDRL. The simulation results show that the ILUPDRL achieves up to 67.53% energy savings while considering the water comfort and water hygiene of occupants.

A Customized Deep Sleep Recommender System Using Hybrid Deep Learning.

This paper proposes a recommendation system based on a hybrid learning approach for a personal deep sleep service, called the Customized Deep Sleep Recommender System (CDSRS). Sleep is one of the most important factors for human life in modern society. Optimal sleep contributes to increasing work efficiency and controlling overall well-being. Therefore, a sleep recommendation service is considered a necessary service for modern individuals. Accurate sleep analysis and data are required to provide such a personalized sleep service. However, given the variations in sleep patterns between individuals, there is currently no international standard for sleep. Additionally, service platforms face a cold start problem when dealing with new users. To address these challenges, this study utilizes K-means clustering analysis to define sleep patterns and employs a hybrid learning algorithm to evaluate recommendations by combining user-based and collaborative filtering methods. It also incorporates feedback top-N classification processing for user profile learning and recommendations. The behavior of the study model is as follows. Using personal information received through mobile devices and data, such as snoring, sleep time, movement, and noise collected through AI motion beds, we recommend sleep and receive user evaluations of recommended sleep. This assessment reconstructs the profile and, finally, makes recommendations using top-N classification. The experimental results were evaluated using two absolute error measurement methods: mean squared error (MSE) and mean absolute percentage error (MAPE). The research results regarding the hybrid learning methods show 13.2% fewer errors than collaborative filtering (CF) and 10.2% fewer errors than content-based filtering (CBF) on an MSE basis. According to the MAPE, the methods are 14.7% more accurate than the CF model and 9.2% better than the CBF model. These results demonstrate that CDSRS systems can provide more accurate recommendations and customized sleep services to users than CF, CBF, and combination models. As a result, CDSRS, a hybrid learning method, can better reflect a user's evaluation than traditional methods and can increase the accuracy of recommendations as the number of users increases.

Discriminate2Rec: Negation-based dynamic discriminative interest-based preference learning for semantics-aware content-based recommendation

Content-based recommender systems (CBRSs) have shown to be highly effective for enhancing user experience in various problems and application domains. Most existing CBRSs employ static preference learning techniques, neglecting that user preferences constantly change over time and are attributed to only a set of certain attributes at a certain period of time due to the existence of temporal dynamics, as only certain attributes of highly preferred items actually reflect user positive interests, while many others might have neutral or even negative influence on user preferences regardless their semantic similarity. Lacking the efficiency in handling these problems poses semantical and temporal incoherence in user profiles, which significantly affects the recommendation accuracy. This paper presents Discriminate2Rec, a three-stage preference learning model that discriminates between items’ attributes based on their influence on user temporal preferences to improve temporal and semantical attribute-level profile coherence for more accurate recommendation. Evaluation is made on three real-world data sets. The results demonstrate the effectiveness of Discriminate2Rec in outperforming state-of-the-art approaches in terms of recommendation accuracy. More in detail, the results also demonstrate that: (1) learning user profiles based on the discrimination between items’ attributes improves the attribute-level profile coherence; (2) modelling the negation in user temporal preferences allows Discriminate2Rec to recommend items whose content are semantically similar to user positive interests to which user negative interests are semantically irrelevant; (3) this novel combination significantly improves temporal and semantical attribute-level profile coherence, and maximises the utilisation of signal, thus significantly improves the recommendation accuracy consistently over time.

Multimedia Recommendation System for Video Game Based on High-Level Visual Semantic Features

The increase of multimedia content in e-commerce and entertainment services creates a new research gap in the field of recommendation systems. The main emphasis of the presented work is on increasing the accuracy of multimedia recommendations using visual semantic content. Recent approaches have shown that the inclusion of visual information is helpful to understand the semantic features for a recommendation model. The researchers have contributed to the field of multimedia item recommendations using low-level visual semantic features. Here, we seek to extend this contribution by exploring the high-level visual semantic content using constant visual attributes for video game recommendation systems. With the exponential growth of multimedia content in the video game industry in the last decade, researchers investigate the importance of personalized video game recommendation techniques. Previous methods have not investigated the importance of visual semantic content for video game recommendations. A practical recommendation system for video games is challenging due to the data diversity, level of user interest, and semantic complexity of features involved. This study proposed a novel method named Deep Visual Semantic Multimedia Recommendation Systems (D_VSMR) to deal with high-level visual features for multimedia recommendation systems. A visual semantic-based video game recommendation system utilizing deep learning methods for visual content learning and user profile learning is introduced. The proposed approach employs content-based techniques to expand users’ profiles. The user profile expansion is based on the visual content of games. The required datasets have been obtained from video game e-commerce platforms like Google Play Store and Amazon for evaluation purposes. The evaluation results have shown that the proposed approach’s accuracy and effectiveness have been improved up to 95.87% compared to the other state-of-the-art methods.

A Framework for Feature Selection Using Natural Language Processing for User Profile Learning for Recommendations of Healthcare-Related Content

This paper presents the work done on recommendations of healthcare related journal papers by understanding the semantics of terms from the papers referred by users in past. In other words, user profiles based on user interest within the healthcare domain are constructed from the kind of journal papers read by the users. Multiple user profiles are constructed for each user based on different categories of papers read by the users. The proposed approach goes to the granular level of extrinsic and intrinsic relationship between terms and clusters highly semantically related relevant domain terms where each cluster represents a user interest area. The semantic analysis of terms is done starting from co-occurrence analysis to extract the intra-couplings between terms and then the inter-couplings are extracted from the intra-couplings and then finally clusters of highly related terms are formed. The experiments showed improved precision for the proposed approach as compared to the state-of-the-art technique with a mean reciprocal rank of 0.76.

Combining novelty and popularity on personalised recommendations via user profile learning

Recommender systems have been widely used by large companies in the e-commerce segment as aid tools in the search for relevant contents according to the user’s particular preferences. A wide variety of algorithms have been proposed in the literature aiming at improving the process of generating recommendations; in particular, a collaborative, diffusion-based hybrid algorithm has been proposed in the literature to solve the problem of sparse data, which affects the quality of recommendations. This algorithm was the basis for several others that effectively solved the sparse data problem. However, this family of algorithms does not differentiate users according to their profiles. In this paper, a new algorithm is proposed for learning the user profile and, consequently, generating personalised recommendations through diffusion, combining novelty with the popularity of items. Experiments performed in well-known datasets show that the results of the proposed algorithm outperform those from both diffusion-based hybrid algorithm and traditional collaborative filtering algorithm, in the same settings.

Dual memory network model for sentiment analysis of review text

In sentiment analysis of product reviews, both user and product information are proven to be useful. Current works handle user profile and product information in a unified model which may not be able to learn salient features of users and products effectively. In this work, we propose a dual user and product memory network (DUPMN) model to learn user profiles and product information for reviews classification using separate memory networks. Then, the two representations are used jointly for sentiment analysis. The use of separate models aims to capture user profiles and product information more effectively. Comparing with state-of-the-art unified prediction models, evaluations on three benchmark datasets (IMDB, Yelp13, and Yelp14) show that our dual learning model gives performance gain of 0.6%, 1.2%, and 0.9%, respectively. The improvements are also deemed very significant measured by p-values.

TweetFit: Fusing Multiple Social Media and Sensor Data for Wellness Profile Learning

Wellness is a widely popular concept that is commonly applied to fitness and self-help products or services. Inference of personal wellness-related attributes, such as body mass index or diseases tendency, as well as understanding of global dependencies between wellness attributes and users' behavior is of crucial importance to various applications in personal and public wellness domains. Meanwhile, the emergence of social media platforms and wearable sensors makes it feasible to perform wellness profiling for users from multiple perspectives. However, research efforts on wellness profiling and integration of social media and sensor data are relatively sparse, and this study represents one of the first attempts in this direction. Specifically, to infer personal wellness attributes, we proposed multi-source individual user profile learning framework named "TweetFit". "TweetFit" can handle data incompleteness and perform wellness attributes inference from sensor and social media data simultaneously. Our experimental results show that the integration of the data from sensors and multiple social media sources can substantially boost the wellness profiling performance.

Concept-based item representations for a cross-lingual content-based recommendation process

The growth of the Web is the most influential factor that contributes to the increasing importance of text retrieval and filtering systems. On one hand, the Web is becoming more and more multilingual, and on the other hand users themselves are becoming increasingly polyglot. In this context, platforms for intelligent information access as search engines or recommender systems need to evolve to deal with this increasing amount of multilingual information. This paper proposes a content-based recommender system able to generate cross-lingual recommendations. The idea is to exploit user preferences learned in a given language, to suggest item in another language. The main intuition behind the work is that, differently from keywords which are inherently language dependent, concepts are stable across different languages, allowing to deal with multilingual and cross-lingual scenarios. We propose four knowledge-based strategies to build concept-based representation of items, by relying on the knowledge contained in two knowledge sources, i.e. Wikipedia and BabelNet. We learn user profiles by leveraging the different concept-based representations, in order to define a cross-lingual recommendation process. The empirical evaluation carried out on two state of the art datasets, DBbook and Movielens, shows that concept-based approaches are suitable to provide cross-lingual recommendations, even though there is not a clear advantage of using one of the different proposed representations. However, it emerges that most of the times the approaches based on BabelNet outperform those based on Wikipedia, which clearly shows the advantage of using a native multilingual knowledge source.

Neural Network Classification for user Profile Learning over Digital Library Recommendation Engine

Objectives: To propose a hybrid recommendation engine to make perfect order of recommendations for online digital library portals. Methods/Statistical Analysis: The proposed model combines the content-based learning based upon the neutralized factors along with the collaborative learning for producing the recommendations on the basis of the interuser similarity, which is learned from the user’s profile representation vector. The hybrid recommendation algorithm is intended to produce more relevant recommendations for the digital library users. Findings: The proposed has been tested over the different sizes of the data. The proposed model has been designed for the discovery of the similar entities using the neural network classification and the top five entities are utilized for the calculation of the missing profile values to compute the recommendations over the digital library systems. The proposed model’s performance has been analyzed and studied using the statistical performance measuring parameters which elaborates the system performance from various perspectives. The experimental results have shows the robust performance of the proposed model in the terms of mentioned parameters. Application/Improvements: The proposed model has been improved the recommendations by the means of the calculating the inter user relevance to evaluate the recommendations for the user in the scope.

A Novel User Profile Learning Approach with Fuzzy Constraint for News Retrieval

This study conducts a novel user profile learning approach based on fuzzy constraints. From the vantage of knowledge representation, a fuzzy constraint network is used not only to present the ambiguity of concepts and the diversity between concepts but also to express a single user profile with dependent multisubjects of interest. From the vantage of problem solving, the construction of a user profile is viewed as a problem of fuzzy constraint satisfaction. The subject of interest is extracted by a spreading activation model. To achieve the information filtering of the retrieved data, fuzzy information gain is employed to reduce unnecessary user feedback for matching the user's retrieval requirements.

A New Profile Learning Model for Recommendation System based on Machine Learning Technique

Recommender systems (RSs) have been used to successfully address the information overload problem by providing personalized and targeted recommendations to the end users. RSs are software tools and techniques providing suggestions for items to be of use to a user, hence, they typically apply techniques and methodologies from Data Mining. The main contribution of this paper is to introduce a new user profile learning model to promote the recommendation accuracy of vertical recommendation systems. The proposed profile learning model employs the vertical classifier that has been used in multi classification module of the Intelligent. Adaptive Vertical Recommendation (IAVR) system to discover the user’s area of interest, and then build the user’s profile accordingly. Experimental results have proven the effectiveness of the proposed profile learning model, which accordingly will promote the recommendation accuracy

A New Profile Learning Model for Recommendation System based on Machine Learning Technique

Recommender systems (RSs) have been used to successfully address the information overload problem by providing personalized and targeted recommendations to the end users. RSs are software tools and techniques providing suggestions for items to be of use to a user, hence, they typically apply techniques and methodologies from Data Mining. The main contribution of this paper is to introduce a new user profile learning model to promote the recommendation accuracy of vertical recommendation systems. The proposed profile learning model employs the vertical classifier that has been used in multi classification module of the Intelligent. Adaptive Vertical Recommendation (IAVR) system to discover the user’s area of interest, and then build the user’s profile accordingly. Experimental results have proven the effectiveness of the proposed profile learning model, which accordingly will promote the recommendation accuracy

Semantics and Agents Oriented Web Personalization

Advent of technologies like semantic web, multi-agent systems, web mining has changed the internet as knowledge provider. Web personalization offers a solution to the information overload problem in current web by providing users a personalized experience, considering their interest, behavior, context and emotions. Semantic web technology is based on use of software agents, ontologies and reasoning to add meaning to web information. An important technology for achieving personalization is the use of independent intelligent software agents. This work reviews, web personalization in the light of semantic web and software agent technology. A comparative study of recent works in the domain of web personalization has been carried out for this purpose. This review highlights ample scope for application of intelligent agents in the web personalization domain for solving many existing issues like personalized content management, user profile learning, modeling and adaptive interactions with users.

A personalized query expansion approach for engineering document retrieval

Engineers create engineering documents with their own terminologies, and want to search existing engineering documents quickly and accurately during a product development process. Keyword-based search methods have been widely used due to their ease of use, but their search accuracy has been often problematic because of the semantic ambiguity of terminologies in engineering documents and queries. The semantic ambiguity can be alleviated by using a domain ontology. Also, if queries are expanded to incorporate the engineer’s personalized information needs, the accuracy of the search result would be improved. Therefore, we propose a framework to search engineering documents with less semantic ambiguity and more focus on each engineer’s personalized information needs. The framework includes four processes: (1) developing a domain ontology, (2) indexing engineering documents, (3) learning user profiles, and (4) performing personalized query expansion and retrieval. A domain ontology is developed based on product structure information and engineering documents. Using the domain ontology, terminologies in documents are disambiguated and indexed. Also, a user profile is generated from the domain ontology. By user profile learning, user’s interests are captured from the relevant documents. During a personalized query expansion process, the learned user profile is used to reflect user’s interests. Simultaneously, user’s searching intent, which is implicitly inferred from the user’s task context, is also considered. To retrieve relevant documents, an expanded query in which both user’s interests and intents are reflected is then matched against the document collection. The experimental results show that the proposed approach can substantially outperform both the keyword-based approach and the existing query expansion method in retrieving engineering documents. Reflecting a user’s information needs precisely has been identified to be the most important factor underlying this notable improvement.

Automatic ontology-based user profile learning from heterogeneous web resources in a big data context

The Web has developed to the biggest source of information and entertainment in the world. By its size, its adaptability and flexibility, it challenged our current paradigms on information sharing in several areas. By offering everybody the opportunity to release own contents in a fast and cheap way, the Web already led to a revolution of the traditional publishing world and just now, it commences to change the perspective on advertisements. With the possibility to adapt the contents displayed on a page dynamically based on the viewer's context, campaigns launched to target rough customer groups will become an element of the past. However, this new ecosystem, that relates advertisements with the user, heavily relies on the quality of the underlying user profile. This profile has to be able to model any combination of user characteristics, the relations between its composing elements and the uncertainty that stems from the automated processing of real-world data. The work at hand describes the beginnings of a PhD project that aims to tackle those issues using a combination of data analysis, ontology engineering and processing of big data resources provided by an industrial partner. The final goal is to automatically construct and populate a profile ontology for each user identified by the system. This allows to associate these users to high-value audience segments in order to drive digital marketing.

Intelligent Information Retrieval and Recommender System Framework

With huge growth in enterprise data, intelligent information retrieval methods have gained research focus. This paper addresses the difficulty of retrieving relevant, pertinent, and novel information for a large system that involves fusion of data in different formats such as, text, barcode, and images. We propose a framework to combine an intelligent image retrieval and intelligent information retrieval (IIR) along with the user profile learning to develop a recommender system. We demonstrate the application of our proposed framework in a real-life situation.

Semantic Learning Service Personalized

Abstract To provide users with more suitable and personalized service, personalization is widely used in various fields. Current e-Learning systems search for learning resources using information search technology, based on the keywords that selected or inputted by the user. Due to lack of semantic analysis for keywords and exploring the user contexts, the system cannot provide a good learning experiment. In this paper, we defined the concept and characteristic of the personalized learning service, and proposed a semantic learning service personalized framework. Moreover, we made full use of semantic technology, using ontologies to represent the learning contents and user profile, mining and utilizing the friendship and membership of the social relationship to construct the user social relationship profile, and improved the collaboration filtering algorithm to recommend personalized learning resources for users. The results of the empirical evaluation show that the approach is effectiveness in augmenting ...

Ontology-based user profile learning

Personal agents gather information about users in a user profile. In this work, we propose a novel ontology-based user profile learning. Particularly, we aim to learn context-enriched user profiles using data mining techniques and ontologies. We are interested in knowing to what extent data mining techniques can be used for user profile generation, and how to utilize ontologies for user profile improvement. The objective is to semantically enrich a user profile with contextual information by using association rules, Bayesian networks and ontologies in order to improve agent performance. At runtime, we learn which the relevant contexts to the user are based on the user's behavior observation. Then, we represent the relevant contexts learnt as ontology segments. The encouraging experimental results show the usefulness of including semantics into a user profile as well as the advantages of integrating agents and data mining using ontologies.

Designing Websites to Learn User Profiles Quickly

Personalization and recommendation systems require knowledge about the users (user profiles), in order to be able to target products, promotions and advertisements. The faster the profiles are learnt, the sooner the site can start benefiting from these systems. We study how a site can learn the profile of a user as quickly as possible as the user traverses the site.