Similarity Algorithm Research Articles

Patient similarity and other artificial intelligence machine learning algorithms in clinical decision aid for shared decision-making in the Prevention of Cardiovascular Toxicity (PACT): a feasibility trial design

BackgroundThe many improvements in cancer therapies have led to an increased number of survivors, which comes with a greater risk of consequent/subsequent cardiovascular disease. Identifying effective management strategies that can mitigate this risk of cardiovascular complications is vital. Therefore, developing computer-driven and personalized clinical decision aid interventions that can provide early detection of patients at risk, stratify that risk, and recommend specific cardio-oncology management guidelines and expert consensus recommendations is critically important.ObjectivesTo assess the feasibility, acceptability, and utility of the use of an artificial intelligence (AI)-powered clinical decision aid tool in shared decision making between the cancer survivor patient and the cardiologist regarding prevention of cardiovascular disease.DesignThis is a single-center, double-arm, open-label, randomized interventional feasibility study. Our cardio-oncology cohort of > 4000 individuals from our Clinical Research Data Warehouse will be queried to identify at least 200 adult cancer survivors who meet the eligibility criteria. Study participants will be randomized into either the Clinical Decision Aid Group (where patients will use the clinical decision aid in addition to current practice) or the Control Group (current practice). The primary endpoint of this study is to assess for each patient encounter whether cardiovascular medications and imaging pursued were consistent with current medical society recommendations. Additionally, the perceptions of using the clinical decision tool will be evaluated based on patient and physician feedback through surveys and focus groups.SummaryThis trial will determine whether a clinical decision aid tool improves cancer survivors’ medication use and imaging surveillance recommendations aligned with current medical guidelines.Trial registrationClinicalTrials.Gov Identifier: NCT05377320

Link and edge weight prediction in air transport networks — An RNN approach

Predicting the future structure of air transport networks is important for several stakeholders in terms of e.g., access to markets, prospects for economic integration and development of regions. Link and edge weight prediction aims to foretell whether two airports will be connected by a direct flight in a future stage of the development of a network and the frequency with which services will be provided. This work assesses the capacity of popular similarity-based algorithms to predict network evolution in air transport. It also proposes a supervised recurrent neural network-based learning framework (RNN) for link prediction. It draws on a set of topological, temporal and content-based features. Experimental results from network data that maps the European Air Transport Network in the period between 2010 and 2019 show that similarity-based algorithms are not able to predict future network stages well. Their performance in predicting newly emerging links remains below expectations formulated in the earlier link prediction literature. The proposed RNN framework outperforms traditional similarity algorithms by a substantial margin. However, the results suggest that link and edge weight prediction remain challenging in sparse air transport networks. Predictive performance must be optimised even further before forecasts can be used to inform concrete policy decisions.

Evolutionary Image Registration: A Review.

Image registration is one of the most important image processing tools enabling recognition, classification, detection and other analysis tasks. Registration methods are used to solve a large variety of real-world problems, including remote sensing, computer vision, geophysics, medical image analysis, surveillance, and so on. In the last few years, nature-inspired algorithms and metaheuristics have been successfully used to address the image registration problem, becoming a solid alternative for direct optimization methods. The aim of this paper is to investigate and summarize a series of state-of-the-art works reporting evolutionary-based registration methods. The papers were selected using the PRISMA 2020 method. The reported algorithms are reviewed and compared in terms of evolutionary components, fitness function, image similarity measures and algorithm accuracy indexes used in the alignment process.

Hybrid Approach of Fractional Generalized Pareto Motion and Cosine Similarity Hidden Markov Model for Solar Radiation Forecasting

Power from solar energy is not reliable, due to weather-related factors, which diminishes the power system’s reliability. Therefore, this study suggests a way to predict the intensity of solar irradiance using various statistical algorithms and artificial intelligence. In particular, we suggest the use of a hybrid predictive model, combining statistical properties and historical data training. In order to evaluate the maximum prediction steps of solar irradiance, the maximum Lyapunov exponent was applied. Then, we used the cosine similarity algorithm in the hidden Markov model for the initial prediction. The combination of the Hurst exponent and tail parameter revealed the self-similarity and long-range dependence of the fractional generalized Pareto motion, which enabled us to consider the iterative predictive model. The initial prediction was substituted into a stochastic differential equation to achieve the final prediction, which prevents error propagation. The effectiveness of the hybrid model was demonstrated in the case study.

Detection of Burnt Areas by Remote Sensing Techniques: Antalya Manavgat Forest Fire

In this study, the forest fire that occurred in Manavgat district of Antalya on 28 July 2021 and lasted for 15 days was analysed by remote sensing techniques using Landsat 8 satellite images. Satellite images of the study area dated July 2021 before the forest fire and August 2021 after the forest fire were obtained. Burnt areas were identified using data’s such as Normalized Burned Ratio (NBR) and Normalized Vegetation Index (NDVI) indices and Difference Normalized Burned Ratio (DNBR) and Difference Normalized Vegetation Index obtained by using the differences of these indices. The maximum similarity algorithm of pixel-based controlled classification was also applied to the data set. The area destroyed by burning after the forest fire was tried to be calculated with these two indexes. It was investigated whether the results of three different methods were compatible and consistent with the results of the General Directorate of Forestry. Although there are differences between the results, it was determined that the selected method and the materials used were suitable for such studies.

Автоматическое определение сходства Javadoc-комментариев

Code comments are an essential part of software documentation. Many software projects suffer the problem of low-quality comments that are often produced by copy-paste. In case of similar methods, classes, etc. copy-pasted comments with minor modifications are justified. However, in many cases this approach leads to degraded documentation quality and, subsequently, to problematic maintenance and development of the project. In this study, we address the problem of near-duplicate code comments detection, which can potentially improve software documentation. We have conducted a thorough evaluation of traditional string similarity metrics and modern machine learning methods. In our experiment, we use a collection of Javadoc comments from four industrial open-source Java projects. We have found out that LCS (Longest Common Subsequence) is the best similarity algorithm taking into account both quality (Precision 94%, Recall 74%) and performance.

Crystallographic variant mapping using precession electron diffraction data

In this work, we developed three methods to map crystallographic variants of samples at the nanoscale by analyzing precession electron diffraction data using a high-temperature shape memory alloy and a VO2 thin film on sapphire as the model systems. The three methods are (I) a user-selecting-reference pattern approach, (II) an algorithm-selecting-reference-pattern approach, and (III) a k-means approach. In the first two approaches, Euclidean distance, Cosine, and Structural Similarity (SSIM) algorithms were assessed for the diffraction pattern similarity quantification. We demonstrated that the Euclidean distance and SSIM methods outperform the Cosine algorithm. We further revealed that the random noise in the diffraction data can dramatically affect similarity quantification. Denoising processes could improve the crystallographic mapping quality. With the three methods mentioned above, we were able to map the crystallographic variants in different materials systems, thus enabling fast variant number quantification and clear variant distribution visualization. The advantages and disadvantages of each approach are also discussed. We expect these methods to benefit researchers who work on martensitic materials, in which the variant information is critical to understand their properties and functionalities.

Retracted: Research and Development of User Clustering-Based Content Similarity Algorithms in Dance-Assisted Choreography Techniques.

[This retracts the article DOI: 10.1155/2022/1364835.].

Exploring the Tradeoff between Utility and Privacy for Analyzing Stock Trend

In this study of exploring the tradeoff between utility and privacy, we have used the S & P 500 data as a proof of concept, for Association Rule Data Mining using privacy-preserved data. The purpose of choosing S & P 500 was that unlike synthetically generated data, S & P 500 index though public, was real-world data nonetheless and hence incorporated in it all the political and social events that occurred over the period of time under consideration. Secondly, we could use stock price as a quasi-identifier. Any seasoned stockbroker would be able to identify the organization simply by looking at the stock price fluctuations over a period even if the ticker information is removed. And stock price is the most important variable used in any data mining related with stock market. Hence, we had to perturb this one variable in order to test for anonymity and utility. Finally, the data being numerical in nature helped us utilize the statistical data perturbation techniques. Association rule mining techniques were first applied to the original data. The data was then perturbed and, once again, using binary similarity algorithms, trend matching was performed on the identified stocks from the original data. The three tests of different random data generations were evaluated for data utility and privacy. The results re-emphasized on the tradeoff between the data utility and privacy where greater perturbation meant greater privacy but higher utility loss and vice versa.

Implementasi Natural Language Processing (NLP) dan Algoritma Cosine Similarity dalam Penilaian Ujian Esai Otomatis

Evaluation of learning is an activity that is routinely carried out in the lecture process. The essay exam is a test in the form of questions that aim so that the answers given are in the form of descriptions based on student’s understanding in accordance with what they know. The results of the various answers are a separate consideration in correcting whether the answer is in accordance with the answer key or not. This resulted in each question on the essay exam having its own weight which would later be added up cumulatively to get a total score. This study implements Natural Language Processing (NLP) and Cosine Similarity algorithms to automatically assess essay exams. Document Similarity is one of the tasks in Natural Language Processing (NLP) to check the degree of document similarity. The algorithm used to check the level of similarity is Cosine Similarity which uses two vectors to measure the degree of similarity of documents with the results ranging from 0 to 1. Processing student answer data for three essay questions gets the expected results. The results of the Cosine Similarity calculation in question no 1 show that M3 students have answers with a similarity level of 90.58%. Whereas for question no 2 M1 students had answers with a similarity level of 87.71% and finally for question no 3 M1 students had answers with a similarity level of 76.70%.

Case-driven assessment of power grid emergency response capability considering typhoon disaster scenario diversity

The effective assessment of power grid emergency response capability (PERC) is the most fundamental aspect in the emergency continuity plans for urban power grid protection. Existing PERC assessment methods assign PERC indicator values under specific disaster scenarios based on one-round expert scores. However, the emergency continuity plans are often continuous and consider diverse scenarios, which makes the assessment quite time-consuming and unrealistic given too many to consider. In this paper, a case-driven method is proposed to solve the assessment problem of PERC with multiple typhoon disaster scenarios while considering the past experience as a guide for real decision making. Firstly, in terms of hierarchical PERC indicator framework, the <typhoon scenario - response indicator - assessment result > framework structure of power grid emergency cases is proposed to make case experience modeled and reusable. Secondly, a case scenario similarity algorithm is applied to select the PERC assessment rules (by mapping relations between response indicators and assessment results) that are most similar to the real scenarios. Then the rules are integrated while considering rule variety and case weights. Next, the hierarchical assessment results are obtained by integrating the confidence coefficient values and corresponding utility values. Finally, the method is verified based on the empirical data from Shenzhen, China. The results show the proposed method not only helps in integrating case experience but also assists in quickly generating reasonable PERC assessment results.

Implementation of the item-based collaborative filtering method on a web-based culinary tourism recommendation system (case study: Magelang City)

Culinary is a type of business that will always exist, this is because one of the factors of human needs is in the form of food. Therefore it is necessary to have a system that provides recommendations to provide input to consumers to minimize consumer confusion about the many culinary places. In this study, to obtain recommendations, the Item-based Collaborative Filtering method will be used to approach consumers with one another. Meanwhile, to get this approach, the approach that will be assessed is an approach to rating culinary places, to get a consumer rating it will give value to 4 aspects, namely: place value, service value, view value, and speed of service value. This study uses the consine similarity algorithm to get the value of the approach. The results obtained by this study using the Likert scale method got a value of 71.6656% "fulfilled". Meanwhile, in the black-box test, the score is 100% functioning, and in the test to determine the accuracy of manual calculations and the system, the value is 100% the same.

Product Innovation Design Process Model Based on Functional Genes Extraction and Construction

The product innovation design is an innovation-intensive process that requires abundant knowledge and design experience, even across industries and disciplines, as a resource for problem solving. Therefore, designers need to have a good command of more design knowledge in this process. However, due to varied expressions of design knowledge (function, principle, structure) in different fields lead to the difficulty in representation of design knowledge, which makes designers unable to innovate by directly using design knowledge. Therefore, to solve the problem above, this paper, analogous to biological gene, proposes a product innovation process model based on functional gene extraction and construction. The proposed process model normalizes design knowledge through functional genes and help designers acquire design knowledge in different fields, which further promotes the product innovation design. The process model consists of four following steps. First, Obtain the overall function of product based on user needs and decompose it. Second, Build the functional gene model based on digital twin ones to expand the breadth of design knowledge and facilitate the retrieval. Third, Screen similar functional gene by the functional similarity algorithm. Finally, Obtain the design scheme through recombination, transcription, and translation of the functional gene. The feasibility of the research method is verified by using a shared bicycle parking device as an example.

LncRNA–disease association prediction method based on the nearest neighbor matrix completion model

State-of-the-art medical studies proved that long noncoding ribonucleic acids (lncRNAs) are closely related to various diseases. However, their large-scale detection in biological experiments is problematic and expensive. To aid screening and improve the efficiency of biological experiments, this study introduced a prediction model based on the nearest neighbor concept for lncRNA–disease association prediction. We used a new similarity algorithm in the model that fused potential associations. The experimental validation of the proposed algorithm proved its superiority over the available Cosine, Pearson, and Jaccard similarity algorithms. Satisfactory results in the comparative leave-one-out cross-validation test (with AUC = 0.96) confirmed its excellent predictive performance. Finally, the proposed model’s reliability was confirmed by performing predictions using a new dataset, yielding AUC = 0.92.

Varieties of corona news: a cross-national study on the foundations of online misinformation production during the COVID-19 pandemic.

The online version contains supplementary material available at 10.1007/s42001-022-00193-5.

Hierarchical concept Bottleneck models for vision and their application to explainable fine classification and tracking

Improving the explainability of Computer Vision models based on Deep Learning has recently become a compelling problem, ensuring reliable predictions to the end-user and enabling more fine-grained classifications. Recently, Concept Bottleneck models have been proposed for images classification, partitioning the problem in two stages and thereby defining a hierarchy of concepts. So far, however, little work has been done to investigate the applicability of this approach to other datasets with higher intra-class variability and ambiguity, and to discuss its flexibility to tasks different from whole-images classification. In this work we develop and discuss a Concept Bottleneck model for images segmentation, objects fine classification and tracking, and compare it to more classical methods based on Mask R-CNN and images similarity algorithms. All our models are trained and tested on a dataset comprised of pictures of fridges filled with various objects, however the method can be applied to any fine classification task. The proposed model makes full use of the hierarchy in concepts, exploiting the relationships between different categories at the same hierarchical level and relying on a novel method for handling multi-labels classifications. We show that the performance on fine classification is on par with a regular Mask R-CNN, but with a significant increase in explainability and in handling classes confusion. New explainable metrics are proposed to quantitatively evaluate the increase in explainability. We also demonstrate the effectiveness of the derived Concept Bottleneck features on related tasks, i.e., the tracking of objects between consecutive pictures in a sequence. The code is released as open source and available at https://opensource.silicon-austria.com/pittinof/hierarchical-concept-bottleneck.

Research on stratigraphic correlation method based on dynamic time bending distance algorithm

Stratigraphic division and correlation are the basis of oil and gas exploration. With the increasing degree of oilfield encryption, the traditional artificial comparison method is difficult to meet the actual needs of oilfield development, and the comparison method of different people is also subjective and different problems, it is difficult to form a more scientific quantitative method. In this paper, the dynamic time-bending distance algorithm of curve similarity algorithm is used for stratigraphic correlation, and two constraint methods based on beam constraint and parallelogram constraint are used to improve the dynamic time-bending distance algorithm. The application of dynamic time-bending distance algorithm in formation correlation is verified by using the real logging data of Nan8 area of Daqing Oilfield. The experimental results show that the dynamic time-bending distance algorithm improves the efficiency of formation correlation and achieves a good formation correlation effect on the real logging curve.

Text mining: identification of similarity of text documents using hybrid similarity model

The volume of data that are accessible on the internet has increased dramatically. This growth of data will only increase exponentially in the future as more data exhaust devices are connected to the network. A part of these data consists of documents from various sources. As the data from various digital sources increase, it becomes tough to perform the process of identification of relevant information which is most essentially needed for their further usage. The goal of this research is to present a hybrid similarity algorithm that uses text summarization techniques to identify papers that are similar in terms of both semantic and contextual similarity. Some of these methods aim to quantify the corpus’s polysemy quotient using deep learning with numerous layers and prebuilt Natural Language Processing (NPL) models to determine document similarity. In comparison with other conventional algorithms, the experimental results of our model showed an accuracy of 76.25%.

Grouping of orally ingested silica nanomaterials via use of an integrated approach to testing and assessment to streamline risk assessment

BackgroundNanomaterials can exist in different nanoforms (NFs). Their grouping may be supported by the formulation of hypotheses which can be interrogated via integrated approaches to testing and assessment (IATA). IATAs are decision trees that guide the user through tiered testing strategies (TTS) to collect the required evidence needed to accept or reject a grouping hypothesis. In the present paper, we investigated the applicability of IATAs for ingested NFs using a case study that includes different silicon dioxide, SiO2 NFs. Two oral grouping hypotheses addressing local and systemic toxicity were identified relevant for the grouping of these NFs and verified through the application of oral IATAs. Following different Tier 1 and/or Tier 2 in vitro methods of the TTS (i.e., in vitro dissolution, barrier integrity and inflammation assays), we generated the NF datasets. Furthermore, similarity algorithms (e.g., Bayesian method and Cluster analysis) were utilized to identify similarities among the NFs and establish a provisional group(s). The grouping based on Tier 1 and/or Tier 2 testing was analyzed in relation to available Tier 3 in vivo data in order to verify if the read-across was possible and therefore support a grouping decision.ResultsThe measurement of the dissolution rate of the silica NFs in the oro-gastrointestinal tract and in the lysosome identified them as gradually dissolving and biopersistent NFs. For the local toxicity to intestinal epithelium (e.g. cytotoxicity, membrane integrity and inflammation), the biological results of the gastrointestinal tract models indicate that all of the silica NFs were similar with respect to the lack of local toxicity and, therefore, belong to the same group; in vivo data (although limited) confirmed the lack of local toxicity of NFs. For systemic toxicity, Tier 1 data did not identify similarity across the NFs, with results across different decision nodes being inconsistent in providing homogeneous group(s). Moreover, the available Tier 3 in vivo data were also insufficient to support decisions based upon the obtained in vitro results and relating to the toxicity of the tested NFs.ConclusionsThe information generated by the tested oral IATAs can be effectively used for similarity assessment to support a grouping decision upon the application of a hypothesis related to toxicity in the gastrointestinal tract. The IATAs facilitated a structured data analysis and, by means of the expert’s interpretation, supported read-across with the available in vivo data. The IATAs also supported the users in decision making, for example, reducing the testing when the grouping was well supported by the evidence and/or moving forward to advanced testing (e.g., the use of more suitable cellular models or chronic exposure) to improve the confidence level of the data and obtain more focused information.

Sustainable Development: A Semantics-aware Trends for Movies Recommendation System using Modern NLP

Abstract Recommendation systems are an important feature in the proposed virtual life, where users are often stuck with choices most of the time and need help to be able to find what they are looking for. In this work, contentbased techniques have been employed in the proposed recommender system in two ways, a deep review for content and features contents such as (cast, crew, keywords, and genres) has been conducted. A preprocessing stage using TF-IDF and CountVectorizer methods have been employed efficiently to prepare the dataset for any similarity measurements. Cosine similarity algorithm has been employed as well with and without sigmoid and linear kernals. The achieved result proves that similarities between movies using TF-IDF with - Cosine similarity (sigmoid kernel) overcomes the TF-IDF with - Cosine similarity (linear_kernel) and Cosine similarity with CountVectorizer in collaborative filtering. The accuracy values of different machine learning models are validated with K-fold Cross Validator techniques. The performance evaluation has been measured using ROOT Mean Square Error and Mean Absolute Error. Five Machine learning algorithms (NormalPredictor, SVD, KNNBasic (with k=20 and K=10), KNNBasic (with sim_options), and NMF (in several rating scales)). Accuracies are finally been validated with 3 folds from each validator. The best achieved RMSE and MAE scores are using SVD (RMSE = 90%) and (MAE = 69%), followed by KNNBasic (with sim_options, K= 20), NMF, KNNBasic (K=20), KNNBasic (K=10), ending with KNNBasic(sim_options, K= 10). Keywords: Recommendation System, Sustainable Development, Artificial Intelligence, Collaborative Filtering, Content-Based, Cosine Similarity, Movies Recommendation, NLP, Machine Learning Application.