Reasonable Interpretation Research Articles

P-143. Interpreting Prognostic Factors Associated with In-Hospital COVID-19 Mortality in a Peru Sample using Shapley Additive Explanations on Ensemble Learning

Abstract Background COVID-19 has infected over 4.5 million and killed more than 222,000 people in Peru. Research on key factors for in-hospital mortality in these patients is limited. Our study aims to understand prognostic factors linked to the mortality of patients with COVID-19 receiving treatment in Peru. A modified SHapley Additive exPlanations (SHAP) method was proposed. Table 1 Methods Clinical and laboratory data were collected prospectively about all patients that were admitted in the inpatient setting at Hospital De Alta Complejidad Virgen De La Puerta and Hospital víctor Lazarte Echegaray Essalud from March 1st to December 31st 2020 with a COVID-19 diagnosis clinically. Ten base predictors were first implemented using the ML algorithms such as Random Forest. Prediction performance on the area under the receiver operating characteristic (AUC) was computed. The SHAP method was then applied to each predictor based on resampled train sets for 30 trials. Factor importance in each trial was estimated by averaging SHAP scores across all predictors, weighted by AUC. Table 2 (Part 1/3) Results After applying inclusion and exclusion criteria (n=) 1,857 patients were selected from a total of 2,000 admitted patients. Several factors out of 281 in total achieved stable importance scores with statistical significance while others were not. Among poor prognostic factors were low PaO2/FiO2 ratio (SV: 4.135, 95% CI: [4.107, 4.163]), shortness of breath (SV: 4.975, 95% CI: [4.864, 5.086]), and productive cough (SV: 2.037, 95% CI: [1.99, 2.080]). Among favorable factors were high PaO2/FiO2 ratio (SV: 3.832, 95% CI: [3.816, 3.848]), low D dimer (SV: 0.875, 95% CI: [0.859, 0.891]) and lymphocyte counts (SV: 1.319, 95% CI: [1.297, 1.340]). For medications, dexamethasone w. air (SV: 2.186, [2.095, 2.277]) was the most effective. For oxygen support, binasal cannula (SV: 0.841, 95% CI: [0.826, 0.857]) and high flow cannula (SV: 0.663, 95% CI: [0.633, 0.693]) are associated with more favorable prognosis than invasive ones. Table 2 (Part 2/3) Conclusion Clinical symptoms and evidence associating acute respiratory distress and sepsis, as well as immune responses less specific to viruses, suggest a likelihood of COVID-19 mortality, consistent with global findings. Our SHAP variant generated reasonable prognostic interpretation about COVID-19 mortality in Peru. Table 2 (Part 3/3) Disclosures All Authors: No reported disclosures

Innovative approach of nomography application into an engineering educational context.

Nomography is considered a branch of mathematics introduced by Maurice d'Ocagne in 1884 in France. The past century saw nomography grow as a graphical computing method used by scientists and engineers wishing to solve complex problems to a practical precision. Even though nomography has declined with the introduction of calculators and computers, it still offers potential in an educational setting. The recent development of open-source software is helping promote the use of nomograms among scholars in engineering courses who are aware of nomography's capabilities. The main reason for this apparent and renewed interest in nomography is the capability of open-source software to generate customized and precise nomograms in seconds without the previously required mathematical background. In this work, we introduce Nomogen, a Python package able to build reliable and scalable 3-variable nomograms while avoiding past drawbacks such as manipulating determinants or manually drawing the scales. In this way, some nomograms generated by Nomogen have been tested on undergraduate and graduate students from different engineering backgrounds. Subsequently, a Likert scale survey was conducted, which showed that students had a great and renewed interest in nomography and found it helpful in the engineering learning process. Even though 78.4% of the respondent had never used nomograms, 86.5% believed that these analogical graphs allow a reasonable interpretation of the phenomenon when there are many variables, and, as a result, nomography with the assistance of open-source software, such as Nomogen or PyNomo, should be incorporated in the teaching process as part of their engineering education syllabus.

Forecasting Methanol-to-olefins Product Yields based on Relevance Vector Machine with Hybrid Kernel and Rolling-windows

Light olefins (ethylene and propylene) have become prominent in chemical industries. Forecasting of the yields of light olefins plays a crucial role in monitoring and optimizing the Methanol-to-olefins (MTO) process. In this work, we introduce an approach for forecasting the yields of ethylene and propylene in the MTO process with the Relevance Vector Machine (RVM) model, which is uniquely enhanced with hybrid kernels and a rolling window methodology. Through an in-depth analysis of 32 independent variables and their pairwise differences, our research pinpoints temperature and pressure as the most critical factors influencing the yields of ethylene and propylene, respectively. The model showcases satisfactory predictive accuracy and reasonable interpretability compared with the traditional statistical and popular machine learning models, marking a step forward in the predictive modeling of chemical engineering processes.

THE METHODOLOGICAL SYSTEM FOR GEOMETRIC INTUITION DEVELOPMENT OF PRIMARY SCHOOLCHILDREN

This article examines the methodological system for geometric intuition development of primary schoolchildren and its usage in the process of teaching mathematics in primary grades, which makes the lesson interesting, effective and purposeful, and the solution of assignments becomes more visual and accessible. The content’s theoretical analysis, as well as practically reasonable interpretations are the basis of the research-experimental part of geometric intuition’s methodological system, which enables students to self-education and allows a creative knowledgeable teacher to have flexible control. In particular, the article presents a primary schoolchild’s geometric intuition as a special means of cognition, which is characterized by direct perception of truth, its formation and condition for the development of preliminary knowledge of geometry. In this sense, developing problem of learner’s higher thinking is the basis of modern educational system, and often primary schoolchildren’s geometric intuitive thinking exceeds their arithmetic thinking. However, when studying the content and learning process of teaching preliminary geometric knowledge in primary school, intuition is manifested both as a form, as a method, and as a means of cognition, which is carried out through subconscious processing of information received by learners, aimed at developing of primary schoolchildren’s geometric intuition using assignments’ system. Meanwhile, based on the studied theoretical material, pedagogical and psychological characteristic features and experimental analyses, we implement an innovative author’s lesson.

Preventable diagnostic errors of lower gastrointestinal perforation: a secondary analysis of a large-scale multicenter retrospective study

BackgroundLower gastrointestinal perforation (LGP) is an acute abdominal condition associated with a high mortality rate. Timely and accurate diagnosis is crucial. Nevertheless, a diagnostic delay has been estimated to occur in approximately one-third of the cases, and the factors contributing to this delay are yet to be clearly understood. This study aimed to evaluate the diagnostic process for appropriate clinical reasoning and availability of image interpretation in cases of delayed diagnosis of LGP.MethodsA secondary data analysis of a large multicenter retrospective study was conducted. This descriptive study analyzed data from a multicenter, observational study conducted across nine hospitals in Japan from January 2015 to December 2019. Out of 439 LGP cases, we included 138 cases of delayed diagnosis, excluding patients with traumatic or iatrogenic perforations, or those secondary to mesenteric ischemia, appendicitis, or diverticulitis. Clinical history and computed tomography (CT) imaging information were collected for 138 cases. Additionally, information on the clinical course of 50 cases, which were incorrectly diagnosed as gastroenteritis, constipation, or small bowel obstruction, was also collected.ResultsIn 42 (30.4%) cases of delayed diagnosis of LGP, CT imaging was performed before diagnosis, indicating a missed opportunity for timely diagnosis. Moreover, 33 of the 50 patients initially diagnosed with gastroenteritis, constipation, or small bowel obstruction at the time of initial examination had atypical findings that were not consistent with the initial diagnosis. Of the 138 cases with delayed diagnosis in our study, 67 cases (48.6%) showed problems with either the interpretation of CT scans or with the process of clinical reasoning.ConclusionOur retrospective study results indicate that approximately half of the cases with delayed diagnosis of LGP were due to problems in interpreting CT images or in clinical reasoning. This finding suggests that clinical reasoning and image interpretation by radiologists are important in improving the diagnostic process for LGP.

Использование комплексной дифференциации и кластеризации геолого-промысловой информации при прогнозировании продуктивности залежей Южно-Татарского свода

The problem of reliable calculation and reasonable interpretation of various productivity coefficient forecasting models is complicated by the presence of hidden patterns between geological and field parameters. The object of the study was the carbonate reservoirs of the deposits of the South Tatar arch, the data on which were obtained as a result of generalization and integration of the results of operating more than 700 wells for various purposes. Twenty geological and physical parameters were identified for the study, using which a discriminant analysis was successfully carried out. Multidimensional regression modeling was carried out for three selected groups of homogeneous objects. The simulation results have shown the efficiency of complex differentiation and clustering of geological and field information, which is based on the hierarchical structuring of parameters, which makes it possible, even with a minimum amount of data at a relevant level, to successfully identify the main trends in the influence of model parameters on the productivity coefficient, expanding the scope of their application to solve a range of basic and indirect problems of rational development of oil fields.

Analysis of Ambiguity in International News Headlines

Understanding ambiguity is one thing that will be useful. Ambiguity is not the same as 'vagueness'. Vagueness means that things have no interpretation, while ambiguity has other reasonable interpretations. By understanding ambiguity, you can learn to think about a problem from a new perspective and explain that thought to others. This study contains the ambiguity found in the headline of news on the British Broadcasting Corporation or BBC news television service site. This study aims to find meaning ambiguity and determine the type of ambiguity of meaning contained in news headlines, including structural ambiguity or lexical ambiguity. The method used in this research is qualitative. The sample in this study is 10 ambiguous headlines that contain ambiguity, it is either structural/grammatical ambiguity and lexical ambiguity. The researchers conclude there are still news headlines that might confuse the readers because they can have two interpretations. So, this study displays the explanation of how and why there can be 2 interpretations in the news headlines. And the two interpretations can be included in structural ambiguity or lexical ambiguity.

Социальная онтология либерализма в просвещенческой трактовке разума: индивидуалистичность, сенсуализм, максимизаторство, антитрадиционализм

The article raises the question of the origins of modern liberal ideology, which is characterized, on the one hand, by the loss of connection with reality (the collapse of the “end of history” project), but at the same time by the preservation of a specific stability in everyday consciousness. The hypothesis is that liberalism quite successfully exploits the stake on rationality, authorizing itself to be a representative of reason. In this regard, it is appropriate to trace the formation of ideas about rationality, as they were formed in the New Age and especially in the philosophy of the Enlightenment period. Apparently, their defining features are: individualism, i.e. understanding of reason as precisely the thinking ability of an individual; empiricist bias to take into account exclusively the data of the senses. In addition, the interpretation of reason as a means for the most complete achievement of goals that have a source in human sensuality. Finally, an attack on traditional attitudes of consciousness in the Enlightenment, primarily religious ones, as hindering the independent thinking of individuals, interpreted as a clear and simple way to achieve a happy life.

Improving Hydrological Simulations with a Dynamic Vegetation Parameter Framework

Many hydrological models incorporate vegetation-related parameters to describe hydrological processes more precisely. These parameters should adjust dynamically in response to seasonal changes in vegetation. However, due to limited information or methodological constraints, vegetation-related parameters in hydrological models are often treated as fixed values, which restricts model performance and hinders the accurate representation of hydrological responses to vegetation changes. To address this issue, a vegetation-related dynamic-parameter framework is applied on the Xinanjiang (XAJ) model, which is noted as Eco-XAJ. The dynamic-parameter framework establishes the regression between the Normalized Difference Vegetation Index (NDVI) and the evapotranspiration parameter K. Two routing methods are used in the models, i.e., the unit hydrograph (XAJ-UH and Eco-XAJ-UH) and the Linear Reservoir (XAJ-LR and Eco-XAJ-LR). The original XAJ model and the modified Eco-XAJ model are applied to the Ou River Basin, with detailed comparisons and analyses conducted under various scenarios. The results indicate that the Eco-XAJ model outperforms the original model in long-term discharge simulations, with the NSE increasing from 0.635 of XAJ-UH to 0.647 of Eco-XAJ-UH. The Eco-XAJ model also reduces overestimation and incorrect peak flow simulations during dry seasons, especially in the year 1991. In drought events, the modified model significantly enhances water balance performance. The Eco-XAJ-UH outperforms the XAJ-UH in 9 out of 16 drought events, while the Eco-XAJ-LR outperforms the XAJ-LR in 14 out of 16 drought events. The results demonstrate that the dynamic-parameter model, in regard to vegetation changes, offers more accurate simulations of hydrological processes across different scenarios, and its parameters have reasonable physical interpretations.

Understanding the Mid-Infrared Spectra of Protic Ionic Liquids by Density Functional Theory.

Protic ionic liquids (PILs) are promising candidates as electrolytes for proton exchange polymer membrane fuel cells. In order to optimize their properties, a detailed understanding of the molecular interactions within the bulk and at the electrode-electrolyte interface is needed, which can be obtained by infrared spectra. A prerequisite for extracting information on the molecular structure and inter- or intramolecular interactions from an experimental spectrum is a reasonable interpretation of the observed spectral features. Here, we employed density functional theory to understand the vibration modes of PILs composed of ammonium cations and different counteranions. Different from the previous calculation methods performed on small cluster model systems consisting of isolated species, a periodically repeated system of four ion pairs was used in order to approximate the bulk liquid environment. The computed frequencies and IR intensity match well with the corresponding experimental spectra, allowing for its proper interpretation, especially the characteristic features of the interionic interaction. The presented approach enables accurate computation of a variety of ionic liquid systems in a highly efficient way.

Modus Darwin redux

How should we evaluate Darwin and Wallace's arguments for common ancestry over separate ancestry? Elliott Sober defends a likelihood reconstruction of Darwin's reasoning that he dubs modus Darwin: similarity, therefore common ancestry. One assumption of Sober's approach is that separate ancestors have traits that are probabilistically independent. I motivate an objection to this assumption by appeal to 19th century naturalist alternatives such as those of Geoffroy and Owen. On Geoffroy and Owen's separate ancestry models, the ancestors can have traits that are probabilistically dependent. I then prove a generalization of Sober's approach that allows for similarity matching among traits to favour common ancestry over separate ancestry even when the traits of the separate ancestors are probabilistically dependent. I consider Helgeson's recent criticisms of Sober's approach and his alternative interpretation of Darwin's reasoning: more similar, hence, more recent common ancestry. I defend Sober's approach against Helgeson's objections.

Evidence of palaeoenvironmental and climatic changes from the interpreted radar wave pictures of near surface sediments around the River Nile, Assiut, Egypt

Different palaeoenvironmental features that pose natural geological, environmental, and engineering hazards to human operations occur frequently around the Nile Valley. Moreover, where these features were initially created, their relevance focuses on how the urban communities responded to the processes. So, a ground penetrating radar (GPR) field survey was carried out on different paleoenvironments of Pre-Quaternary and Quaternary sediment around Assiut. Deep and critical analyses of georadar facies were made to obtain clear images of these features with unprecedented resolution. The main objective of this study is to find some reasonable geological interpretations for these features. From this study, it is possible to identify and differentiate these features originating from different geological environments and climatological conditions in arid regions such as those around Assiut. In addition, the study serves as guidelines for environmental management and climatic changes for enhancing knowledge of urban development. Also, the study demonstrates how georadar can be used to create precise images of intricate shallow subsurface anatomy with possible palaeoenvironmental and palaeoclimatic indicators.

국제물품매매거래에서 CISG상 지식재산권 침해 물품 인도에 따른 매도인의 책임에 관한 연구: CISG 자문위원회의 Black Letter Rule을 중심으로

Purpose: This study explores the liability of sellers of goods that infringe Article 42 of the United Nations convention on Contracts for the International Sale of Goods (CISG) based on the official opinion of the CISG Advisory Council. Research design, data, and methodology: The requirements for seller liability under Article 42(1) of the CISG includes the seller's awareness and geographic limitations. This study also examines the requirements for seller immunity under Article 42(2), including the buyer's expectation of awareness and investigation. Results: M ost o f t he e xisting o pinions a nd t he o fficial o pinions o f t he C ISG A dvisory Council seem to be consistent with the responsibility of the seller who delivers goods infringing on intellectual property rights. However, regarding the timing for determining the suitability of rights to the goods, there are areas where opinions are different from those of scholars who view them as a transfer of risk. Conclusions: The CISG Advisory Council's official opinion appears to be a reasonable interpretation of Article 42 taking into account its purpose, which is to protect the interest of the purchaser in the free use of goods that are not subject to third-party rights or claims, the Secretariat's commentary on the Convention's legislative history, and the existing views of scholars. Therefore, practitioners should prepare for related situations by referring to the official opinion of the CISG Advisory Council on Article 42 of the CISG.

Enhanced particle removal ability of two representative nonionic surfactants: A reasonable interpretation based on DFT and coarse-grained molecular dynamics methods

Cleaning of copper surfaces after chemical mechanical polishing is indispensable for removing particle residues, which is a critical step in integrated circuits manufacturing. This paper introduces two nonionic surfactants, coconut oil diethanolamine (CDEA) and fatty alcohol polyoxyethylene ether (AEO-9), into a basic alkaline cleaning solution comprising tetraethyl ammonium hydroxide and lysine to investigate their abilities to enhance particle removal performance. Cleaning performance was measured with scanning electron microscopy, which showed that the two surfactants achieved excellent particle removal efficiency (above 95 %), with CDEA performing slightly better in all tested concentrations. Concerning critical micelle concentration, surface tension tests showed that CDEA (at 1500 ppm) had superior wettability and dispersive capacity compared with AEO-9 (at 2000 ppm) and could prevent particle redeposition more effectively. Using density functional theory, CDEA was found to have more polar groups to form hydrogen bonds with the hydrophilic surface of colloidal silica particles or bond with the copper substrate, explaining CDEA’s ability to lower the surface tension of the cleaning solution and improve the zeta potential between particles and the copper substrate. Finally, based on the Martini forcefield, simulations of coarse-grained molecular dynamics provided valuable theoretical insights into the different dispersion abilities of the two nonionic surfactants that CDEA had a more significant diffusion coefficient (0.073 Å2/ps) and was able to form micelles more efficiently in solution compared with that of AEO-9 (0.066 Å2/ps).

Solute structure effect on polycyclic aromatics separation from fuel oil: Molecular mechanism and experimental insights

AbstractIonic liquids (ILs) are promising solvents for separating aromatics from fuel oils. However, studies for separate polycyclic aromatics with ILs are rare and insufficient, and the impact of solute structure on extraction performance still needs to be determined. In this work, we use 1‐ethyl‐3‐methylimidazolium bis([trifluoromethyl]sulfonyl)imide ([EMIM][NTF2]) as an extractant to separate 1‐methylnaphthalene, quinoline, and benzothiophene from dodecane mixtures. Liquid–liquid equilibrium experiments identified the optimal operating conditions. Nine solute molecules, including five alkanes and four aromatic hydrocarbons, were used to study the relationship between extraction performance and solute structure. Molecular dynamics simulation and quantum chemistry calculations gave a deep insight and reasonable interpretation of the structure‐performance relationship at the molecular level. An industrial‐scale extraction process was proposed. The IL can be easily regenerated using heptane as a back‐extractive solvent. A high‐purity fuel oil with aromatic content below 0.5 wt% is obtained after 8‐stage extraction.

Perspective—Concerning a New Mechanistic Model Toward the Catalytic Ammonia Synthesis

A reasonable catalytic mechanistic model should refer to a widely range of catalytic reaction. We believe that all types of catalytic reaction on heterogeneous catalysis should follow a general mechanism, and it is our opinion that with the hall-filled valence orbitals, the atom of catalysts could convert the reactant into reactive radical and/or support the formation of new chemical bond between two reactants via radical dimerization. In our recent publications this new mechanistic model on the catalytic Fischer-Tropsch reaction (the conversion of CO and H<sub>2</sub> to hydrocarbons), electrochemical hydrogen evolution reactions, and hydrogen combustion in various metal catalysts is discussed, and which seems to provide a reasonable interpretation to those catalytic reactions. In the present work it is discussed that this new mechanistic model is suitable to the Haber-Bosch process (catalytic ammonia synthesis) on various transition metal catalysts, and a reasonable explanation is provided on the catalytic property of various transition metal for ammonia synthesis.

A sequential multiple attribute three-way group decision-making method to heterogeneous MAGDM problems with unknown weight information

As a decision model to depict the human cognitive process, three-way decision (3WD) provides a reasonable semantic interpretation for solving practical multi-attribute group decision-making (MAGDM) problems. At the same time, there often exist different types of information due to the subjective and objective criteria in practical decision-making problems, thus it is necessary to develop some efficient frameworks to deal with the MAGDM problems with heterogeneous information. In this paper, based on 3WD theory and sequential strategy, we propose a sequential multiple attribute three-way group decision-making (SMA3WGD) method for the heterogeneous MAGDM problem with unknown weight information to achieve this goal. At first, we compute the weights of decision-makers by using the consensus degree between each of the decision-makers in a group. Then, we calculate the collective evaluation result of multiple decision-makers by utilizing the mathematical planning method with two objectives of minimum difference and high similarity between group evaluation values and individual evaluation values. Meanwhile, a nested sequence of the set of attributes is constructed according to the attribute values offered by decision-makers that have different degrees of consensus on different attributes. On this basis, a multilevel granular structure for the heterogeneous MAGDM problems is constructed, and then the compute of the cost functions and conditional probability is explored in a single level of granular structure, and the thresholds and three-way decision rules of alternatives are further determined depending on the principle of cost minimization. Furthermore, by using a sequential strategy, we obtain a sequential three-way decision result of the heterogeneous MAGDM problems under the multilevel granular structure. At last, the practicality and advantages of the suggested method are demonstrated via a blockchain service provider selection case study, a parameter sensitivity analysis along with a comparison analysis.

Interpretable multi-graph convolution network integrating spatio-temporal attention and dynamic combination for wind power forecasting

Graph neural networks (GNN) can effectively improve wind power forecasting accuracy due to their ability to capture spatial correlations between wind farms. However, conventional GNN-based forecasting models constructing the graphs from only one specific perspective fail to characterize the spatial correlation patterns comprehensively and accurately. Furthermore, their black-box structure leads to deficient interpretability of the forecasting results, which hampers the quantification of forecasting uncertainty and reduces the credibility of applying these forecasting models in practice. To address the above problems, an interpretable ultra-short-term wind power forecasting method is proposed based on multi-graph convolution network integrating spatio-temporal attention and dynamic graph combination. Four undirected weight graphs are constructed to form a multi-graph scheme by considering various geographical and statistical aspects, and then graph convolution network (GCN) and temporal convolution network (TCN) are used to enhance the forecasting performance by leveraging information in both space and time dimensions. Meanwhile, it consists of temporal attention that deeply mines the temporal dependencies and spatial attention that interprets the influence of wind farms on the forecasts. Additionally, to boost the adaptability and generalization capability of the proposed method, an improved reinforcement learning is elaborately designed to combine the results of different graphs by dynamically optimizing their weights to obtain the final forecasts. Experiment results suggest that the proposed method exhibits superior forecasting accuracy and reasonable interpretability compared with other benchmark methods.

Impact of the New Company Law's Limited Contribution System on Limited Liability Companies

Summarizing the ten years of experience and lessons learned from the operation of the contributed capital system, after three years of deliberation and five years of overhaul, a new change in the Company Law was ushered in. The Company Law of the People's Republic of China (2023 Revision) was voted and passed on December 29, 2023, and came into effect on July 1, 2024. The new Company Law continues to implement the Contributed Capital System, replacing the Fully Contributed System with a Limited Contribution System. It consolidates the positive results of the operation of the Contributed Capital System and amends the deficiencies in the actual operation, better implements the autonomy-based Contributed Capital System, and provides a new path for the development of the Contributed Capital System in the future. In this paper, we will start from the impact of the limited liability company (LLC) under the limited contribution system, and discuss the reasonable interpretation path of the shareholders' interest in the limited contribution system, the company's capital formation system, and the shareholders' responsibility in the light of the revision of the new Company Law.

Footbridge damage detection using smartphone-recorded responses of micromobility and convolutional neural networks

This paper presents a footbridge damage detection and classification framework using smartphone-recorded responses of micromobility and deep learning techniques. Time–frequency representations (TFRs) of scooter vibrations are employed to detect and classify footbridge damage severities using a Two-Dimensional Convolutional Neural Network (2D CNN). A One-Dimensional (1D) CNN using scooter frequency spectra was also investigated for comparison. The effectiveness of the method was verified using a numerical model of scooter-footbridge interactions and field tests on a real footbridge. The results indicated that both CNNs were sensitive to footbridge frequency alterations caused by damage in the numerical simulations. Nevertheless, the performance of the 1D CNN experienced a substantial decline in field tests involving stochastic influencing factors, whereas the accuracy of damage classification using the 2D CNN remained high. Finally, reasonable interpretations for the superior performance of the 2D CNN are provided using Shapley Additive Explanations (SHAP) values.