Explore the vast range of titles available.

Aim/Purpose: This study aims to devise a personalized solution for online healthcare platforms that can alleviate problems arising from information overload and data sparsity by providing personalized healthcare services to patients. The primary focus of this paper is to develop an effective medicine recommendation approach for recommending suitable medications to patients based on their specific medical conditions. Background: With a growing number of people becoming more conscious about their health, there has been a notable increase in the use of online healthcare platforms and e-services as a means of diagnosis. As the internet continues to evolve, these platforms and e-services are expected to play an even more significant role in the future of healthcare. For instance, WebMD and similar platforms offer valuable tools and information to help manage patients’ health, such as searching for medicines based on their medical conditions. Nonetheless, patients often find it arduous and time-consuming to sort through all the available medications to find the ones that match their specific medical conditions. To address this problem, personalized recommender systems have emerged as a practical solution for mitigating the burden of information overload and data sparsity-related issues that are frequently encountered on online healthcare platforms. Methodology: The study utilized a dataset of MC ratings obtained from WebMD, a popular healthcare website. Patients on this website can rate medications based on three criteria, including medication effectiveness, ease of use, and satisfaction, using a scale of 1 to 5. The WebMD MC rating dataset used in this study contains a total of 32,054 ratings provided by 2,136 patients for 845 different medicines. The proposed HSMCCF approach consists of two primary modules: a semantic filtering module and a multi-criteria filtering module. The semantic filtering module is designed to address the issues of data sparsity and new item problems by utilizing a medicine taxonomy that sorts medicines according to medical conditions and makes use of semantic relationships between them. This module identifies the medicines that are most likely to be relevant to patients based on their current medical conditions. The multi-criteria filtering module, on the other hand, enhances the approach’s ability to capture the complexity of patient preferences by considering multiple criteria and preferences through a unique similarity metric that incorporates both distance and structural similarities. This module ensures that patients receive more accurate and personalized medication recommendations. Moreover, a medicine reputation score is employed to ensure that the approach remains effective even when dealing with limited ratings or new items. Overall, the combination of these modules makes the proposed approach more robust and effective in providing personalized medicine recommendations for patients. Contribution: This study addresses the medicine recommendation problem by proposing a novel approach called Hybrid Semantic-based Multi-Criteria Collaborative Filtering (HSMCCF). This approach effectively recommends medications for patients based on their medical conditions and is specifically designed to overcome issues related to data sparsity and new item recommendations that are commonly encountered on online healthcare platforms. The proposed approach addresses data sparsity and new item issues by incorporating a semantic filtering module and a multi-criteria filtering module. The semantic filtering module sorts medicines based on medical conditions and uses semantic relationships to identify relevant ones. The multi-criteria filtering module accurately captures patient preferences and provides precise recommendations using a novel similarity metric. Additionally, a medicine reputation score is also employed to further expand potential neighbors, improving predictive accuracy and coverage, particularly in sparse datasets or new items with few ratings. With the HSMCCF approach, patients can receive more personalized recommendations that are tailored to their unique medical needs and conditions. By leveraging a combination of semantic-based and multi-criteria filtering techniques, the proposed approach can effectively address the challenges associated with medicine recommendations on online healthcare platforms. Findings: The proposed HSMCCF approach demonstrated superior effectiveness compared to benchmark recommendation methods in multi-criteria rating datasets in terms of enhancing both prediction accuracy and coverage while effectively addressing data sparsity and new item challenges. Recommendations for Practitioners: By applying the proposed medicine recommendation approach, practitioners can develop a medicine recommendation system that can be integrated into online healthcare platforms. Patients can then utilize this system to make better-informed decisions regarding the medications that are most suitable for their specific medical conditions. This personalized approach to medication recommendations can ultimately lead to improved patient satisfaction. Recommendation for Researchers: Integrating patient medicine reviews is a promising way for researchers to elevate the proposed medicine recommendation approach. By leveraging patient reviews, the approach can gain a more comprehensive understanding of how certain medications perform for specific medical conditions. Additionally, exploring the relationship between MC-based ratings using an improved aggregation function can potentially enhance the accuracy of medication predictions. This involves analyzing the relationship between different criteria, such as medication effectiveness, ease of use, and satisfaction of the patients, and determining the optimal weighting for each criterion based on patient feedback. A more holistic approach that incorporates patient reviews and an improved aggregation function can enable the proposed medicine recommendation approach to provide more personalized and accurate recommendations to patients. Impact on Society: To mitigate the risk of infection during the COVID-19 pandemic, the promotion of online healthcare services was actively encouraged. This allowed patients to continue accessing care and receiving treatment while adhering to physical distancing guidelines and shielding measures where necessary. As a result, the implementation of personalized healthcare services for patients is expected to be a major disruptive force in healthcare in the coming years. This study proposes a personalized medicine recommendation approach that can effectively address this issue and aid patients in making informed decisions about the medications that are most suitable for their specific medical conditions. Future Research: One way that may enhance the proposed medicine recommendation approach is to incorporate patient medicine reviews. Furthermore, the analysis of MC-based ratings using an improved aggregation function can also potentially enhance the accuracy of medication predictions.

Mooring systems are an integral and sophisticated component of offshore assets and are subject to harsh conditions and cyclic loading. The early detection and characterisation of fatigue crack growth remain a crucial challenge. The scope of the present work was to establish filtering and alarm criteria for different crack growth stages by evaluating the recorded signals and their features. The analysis and definition of parametrical limits, and the correlation of their characteristics with the crack, helped to identify approaches to discriminate between noise, initiation, and growth-related signals. Based on these, a filtering criterion was established, to support the identification of the different growth stages and noise with the aim to provide early warnings of potential damage.

The NOMAD‐UVIS instrument on board the ExoMars Trace Gas Orbiter has been investigating the Martian atmosphere with the occultation technique since April 2018. Here, we analyze almost two Mars Years of ozone vertical distributions acquired at the day‐night terminator. The ozone retrievals proved more difficult than expected due to spurious detections of ozone caused by instrumental effects, high dust content, and very low values of ozone. This led us to compare the results from three different retrieval approaches: (a) an onion peeling method, (b) a full occultation Optimal Estimation Method, and (c) a direct onion peeling method. The three methods produce consistently similar results, especially where ozone densities are higher. The main challenge was to find reliable criteria to exclude spurious detections of O3, and we finally adopted two criteria for filtering: (a) a detection limit, and (b) the Δχ2 criterion. Both criteria exclude spurious O3 values especially near the perihelion (180° < Ls < 340°), where up to 98% of ozone detections are filtered out, in agreement with general circulation models, that expect very low values of ozone in this season. Our agrees well with published analysis of the NOMAD‐UVIS data set, as we confirm the main features observed previously, that is, the high‐altitude ozone peak around 40 km at high latitudes. The filtering approaches are in good agreement with those implemented for the SPICAM/MEx observations and underline the need to evaluate carefully the quality of ozone retrievals in occultations. Key Points We compare three different retrieval codes and different criteria to filter spurious detection of ozone on Mars The filtering criteria demonstrate the ozone detection in perihelion season is mostly spurious in both years The three retrieval methods produce consistent results with a maximum percentage difference of ∼30% for large ozone densities below 50 km

Medical errors associated with medication pose significant threats to patients’ safety, primarily due to the abundance of drug information available on various online healthcare platforms, leading to challenges in identifying relevant drugs. To address this issue, drug recommendation systems have been developed to assist in selecting appropriate medications for specific medical conditions. Collaborative filtering approaches have been widely used to generate personalized recommendations for various applications. They are easy to implement, debug, and provide justifiable reasoning for recommended items, which is not readily accessible in several other recommendation approaches. Regardless of their success, they still need further enhancements to address challenges related to insufficient rating data, such as data sparsity and new item problems. This paper proposes a drug recommendation model that effectively employs drug taxonomy and multi-criteria collaborative filtering to tackle these challenges. Drug taxonomy enhances recommendation quality by offering a more organized and granular representation of drugs, while multi-criteria rating captures the patients’ preferences more accurately, enabling accurate recommendations that better match the patient’s specific preferences. Experiments conducted on a real-world drug multi-criteria rating dataset demonstrate that the proposed model outperforms baseline recommendation approaches in addressing these challenges and improving prediction accuracy and coverage, making it a valuable tool to assist patients in selecting relevant drugs for their specific medical conditions.

Multi-criteria collaborative filtering (MC-CF) presents a possibility to provide accurate recommendations by considering the user preferences in multiple aspects of items. However, scalability and sparsity are two main problems in MC-CF which this paper attempts to solve them using dimensionality reduction and Neuro-Fuzzy techniques. Considering the user behavior about items’ features which is frequently vague, imprecise and subjective, we solve the sparsity problem using Neuro-Fuzzy technique. For the scalability problem, higher order singular value decomposition along with supervised learning (classification) methods is used. Thus, the objective of this paper is to propose a new recommendation model to improve the recommendation quality and predictive accuracy of MC-CF and solve the scalability and alleviate the sparsity problems in the MC-CF. The experimental results of applying these approaches on Yahoo!Movies and TripAdvisor datasets with several comparisons are presented to show the enhancement of MC-CF recommendation quality and predictive accuracy. The experimental results demonstrate that SVM dominates the K-NN and FBNN in improving the MC-CF predictive accuracy evaluated by most broadly popular measurement metrics, F1 and mean absolute error. In addition, the experimental results also demonstrate that the combination of Neuro-Fuzzy and dimensionality reduction techniques remarkably improves the recommendation quality and predictive accuracy of MC-CF in relation to the previous recommendation techniques based on multi-criteria ratings.

Recent years have witnessed a rapid explosion of online information sources about restaurants, and the selection of an appropriate restaurant has become a tedious and time-consuming task. A number of online platforms allow users to share their experiences by rating restaurants based on more than one criterion, such as food, service, and value. For online users who do not have enough information about suitable restaurants, ratings can be decisive factors when choosing a restaurant. Thus, personalized systems such as recommender systems are needed to infer the preferences of each user and then satisfy those preferences. Specifically, multi-criteria recommender systems can utilize the multi-criteria ratings of users to learn their preferences and suggest the most suitable restaurants for them to explore. Accordingly, this paper proposes an effective multi-criteria recommender algorithm for personalized restaurant recommendations. The proposed Hybrid User-Item based Multi-Criteria Collaborative Filtering algorithm exploits users' and items' implicit similarities to eliminate the sparseness of rating information. The experimental results based on three real-word datasets demonstrated the validity of the proposed algorithm concerning prediction accuracy, ranking performance, and prediction coverage, specifically, when dealing with extremely sparse datasets, in relation to other baseline CF-based recommendation algorithms.

The volume of healthcare information available on the internet has exploded in recent years. Nowadays, many online healthcare platforms provide patients with detailed information about doctors. However, one of the most important challenges of such platforms is the lack of personalized services for supporting patients in selecting the best-suited doctors. In particular, it becomes extremely time-consuming and difficult for patients to search through all the available doctors. Recommender systems provide a solution to this problem by helping patients gain access to accommodating personalized services, specifically, finding doctors who match their preferences and needs. This paper proposes a hybrid content-based multi-criteria collaborative filtering approach for helping patients find the best-suited doctors who meet their preferences accurately. The proposed approach exploits multi-criteria decision making, doctor reputation score, and content information of doctors in order to increase the quality of recommendations and reduce the influence of data sparsity. The experimental results based on a real-world healthcare multi-criteria (MC) rating dataset show that the proposed approach works effectively with regard to predictive accuracy and coverage under extreme levels of sparsity.

The employed dictionary plays an important role in sparse representation or sparse coding based image reconstruction and classification, while learning dictionaries from the training data has led to state-of-the-art results in image classification tasks. However, many dictionary learning models exploit only the discriminative information in either the representation coefficients or the representation residual, which limits their performance. In this paper we present a novel dictionary learning method based on the Fisher discrimination criterion. A structured dictionary, whose atoms have correspondences to the subject class labels, is learned, with which not only the representation residual can be used to distinguish different classes, but also the representation coefficients have small within-class scatter and big between-class scatter. The classification scheme associated with the proposed Fisher discrimination dictionary learning (FDDL) model is consequently presented by exploiting the discriminative information in both the representation residual and the representation coefficients. The proposed FDDL model is extensively evaluated on various image datasets, and it shows superior performance to many state-of-the-art dictionary learning methods in a variety of classification tasks.

In data assimilation (DA) schemes for numerical weather prediction (NWP) systems, the estimation of forecast error covariances is a key point to get some flow dependency. As shown in previous studies, ensemble data assimilation methods are the most accurate for this task. However, their huge computational cost raises a strong limitation to the ensemble size. Consequently, covariances estimated with small ensembles are affected by random sampling errors. The aim of this study is to develop a theory of covariance filtering in order to remove most of the sampling noise while keeping the signal of interest and then to use it in the DA scheme of a real NWP system. This first part of a two-part study presents the theoretical aspects of such criteria for optimal filtering based on the merging of the theories of optimal linear filtering and of sample centered moments estimation. Its strength relies on the use of sample estimated quantities and filter output only. These criteria pave the way for new algorithms and interesting applications for NWP. Two of them are detailed here: spatial filtering of variances and covariance localization. Results obtained in an idealized 1D analytical framework are shown for illustration. Applications on real forecast error covariances deduced from ensembles at convective scale are discussed in a companion paper.

Researchers would like to publish their research articles in reputed journals along with quick review time. However, with the growing number of academic publications, it is becoming more difficult for scholars to find venues that are relevant to their domain. This study aims on the development of a technique that focuses on the priorities of the researchers that are linked to the recommendation of suitable suggestion of publication journal. The developed Recommendation System (RS) takes title, abstract, and keyword of the manuscript to be submitted. The proposed algorithm, named PUB-VEN which is hybridization of Content-Based Filtering (CBF), and Collaborative Filtering (CF), which is integrated with the Multi-Criteria Decision Making (MCDM) process to provide suitable journal recommendations by considering the researcher's point of view about different attributes gathered such as impact factor, eigen factor, average review time, etc. which affect the research process effectively. Our results demonstrate that the PUB-VEN provides better recommendations in comparison with state-of-the-art algorithms such as Term Frequency and Inverse Document Frequency (TF-IDF) and Latent Semantic Analysis (LSA). The study concluded that PUB-VEN is providing better precision, recall, F1 Score, Discounted Cumulative Gain (DCG), and Normalized DCG (NCDG). For precision, the gain ranges from 1% to 16%, the improvement in recall is between 33% and 3%, the betterment of result in F1 is by the ratio which ranges from 27% and 2%, the improvement in the result of DCG lies between 15% and 5% and the result of NDCG gain ranges from 6% to 1%. It is useful for the researchers in finding suitable venue for publication.