检索结果-内蒙古大学图书馆

3rd International Symposium on computer Technology and Information science, ISCTIS 2023

作者： Li, Yifan Faculty of Science and Technology Department of Statistics and Data Science Zhuhai China

ISBN: (纸本)9798350325386

The mental health and mental illness crisis has become increasingly acute in recent years, and many digital solutions with artificial intelligence at their core offer hope for reversing the deterioration of our mental health. Machine deep learning techniques can be used to analyse big data to build predictive models for psycho-education, assessment and screening to assess the mental health status of subjects, and can help the clinical community discover information that is not available to many traditional psychological research tools. This paper presents an in-depth analysis of a mental health survey and examines how it can be applied to the Al/ML domain of mental health research and how machine learning models can be used in this domain for fitting and prediction. Based on this, the importance of the presence or absence of current mental health disorders on other characteristics of respondents is assessed and visualised. It was found that the Cross Gradient Booster (Random Forest) model gave the best prediction fit among the various types of machine learning models, and the Grid Search algorithm was used to confirm that the final model had the highest accuracy of 0.79784 at a learning rate of 0.1. The Permutation Importance analysis revealed that the most important characteristic is whether or not the person has suffered from a mental health disorder in the past. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

On Batch Teaching Without Collusion

引用

Journal of Machine Learning Research 2023年第1期24卷 1-33页

作者： Fallat, Shaun Kirkpatrick, David Simon, Hans U. Soltani, Abolghasem Zilles, Sandra Department of Mathematics and Statistics University of Regina Canada Department of Computer Science University of British Columbia Canada Max Planck Institute for Informatics Germany Department of Computer Science University of Regina Canada

Formal models of learning from teachers need to respect certain criteria to avoid collusion. The most commonly accepted notion of collusion-avoidance was proposed by Goldman and Mathias (1996), and various teaching models obeying their criterion have been studied. For each model M and each concept class C, a parameter M-TD(C) refers to the teaching dimension of concept class C in model M—defined to be the number of examples required for teaching a concept, in the worst case over all concepts in C. This paper introduces a new model of teaching, called no-clash teaching, together with the corresponding parameter NCTD(C). No-clash teaching is provably optimal in the strong sense that, given any concept class C and any model M obeying Goldman and Mathias’s collusion-avoidance criterion, one obtains NCTD(C) ≤ M-TD(C). We also study a corresponding notion NCTD+ for the case of learning from positive data only, establish useful bounds on NCTD and NCTD+, and discuss relations of these parameters to other complexity parameters of interest in computational learning theory. We further argue that Goldman and Mathias’s collusion-avoidance criterion may in some settings be too weak in that it admits certain forms of interaction between teacher and learner that could be considered collusion in practice. Therefore, we introduce a strictly stronger notion of collusion-avoidance and demonstrate that the well-studied notion of Preference-based Teaching is optimal among all teaching schemes that are strongly collusion-avoiding on all finite subsets of a given concept class. ©2022 Shaun Fallat, David Kirkpatrick, Hans U. Simon, Abolghasem Soltani, and Sandra Zilles.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

The Beta Exponential Power Series Distribution

引用

Annals of data science 2023年第5期10卷 1157-1178页

作者： Khojastehbakht, Nafiseh Ghatari, Amirhossein Samani, Ehsan Bahrami Department of Statistics Shahid Beheshti University Tehran Iran Department of Statistics Faculty of Mathematic and Computer Science Amirkabir University of Technolgy Tehran Iran

In this paper, we investigate to propose a new statistical distribution based on power series. We introduce a new family of distributions which are constructed based on a latent complementary risk problem and are obtained by compounding Beta Exponential (BE) and Power Series distributions. The new distribution contains, as special sub-models, several important distributions which are discussed in the literature, such as Beta Exponential Poisson (BEP) distribution, Beta Exponential Geometric (BEG) distribution, Beta Exponential Logarithmic (BEL) distribution, Beta Exponential Binomial (BEB) distribution as special cases. The hazard function of the BEPS distributions can be increasing, decreasing or bathtub shaped among others. The comprehensive mathematical properties of the new distribution is provided such as closed-form expressions for the density, cumulative distribution, survival function, failure rate function, the r-th raw moment, maximum likelihood estimation and also the moments of order statistics. The proposed type of distributions is used to modeling simulated and real datasets. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

Leveraging XGBoost and Clinical Attributes for Heart Disease Prediction 5

Leveraging XGBoost and Clinical Attributes for Heart Disease...

引用

5th International Conference on Sustainable Communication Networks and Application, ICSCNA 2024

作者： Susmitha, Kota Bhargavi, Maridu Sri, Achyuta Mohitha Sai Reddy, Bogala Devi Prasaad Satyanarayana, Paladugu Siva Rao, K V Ranga Vignan's Foundation for Science Technology and Research Department of Computer Science and Engineering Guntur India Department of Computer Science and Engineering-Data Science Hyderabad India

ISBN: (纸本)9798331530013

Despite the various advances in medical technology, heart disease continues to rank among the leading causes of mortality in the world, killing millions each year. There is hope that the risks involved with heart disease can be reversed in time and the outcomes of patients improved with earlier prediction and diagnosis. However, the use of conventional methods of diagnostic techniques tends to be very resource-intensive and inaccessible to other more disadvantaged areas. The present work defines a model for machine learning in predicting heart disease and illustrates a more efficient and accessible solution. There is a risk for heart disease that is determined using a set of various clinical features. The algorithm used in the prediction of heart disease is XG Boost that are applied to the dataset. The method provides a reliable tool for the early detection of heart disease, thus helping healthcare providers to improve preventive care and treatment strategies. The results of this study will be helpful in advancing research to identify other avenues for prediction of heart disease. © 2024 IEEE.

关键词： early diagnosis Heart disease UCI (University California Irvine) Heart Disease dataset XG Boost

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Approach to Identify the Incidence of Human Deaths Due to Human-Elephant Conflicts 5

Machine Learning Approach to Identify the Incidence of Human...

引用

5th International Conference on Advanced Research in Computing, ICARC 2025

作者： Lakshitha, W.A.D.M. Chandrasekara, N.V. Dissanayaka, D.M.P.V. Kavinga, H.W.B. University of Kelaniya Department of Statistics & Computer Science Kelaniya Sri Lanka

ISBN: (纸本)9798331530983

Human-Elephant Conflict (HEC) is one of the predominant issues Sri Lanka is coping with, where the North Central Province (NCP) has recorded the highest human deaths due to HEC. However, nowadays, most travelers are cautious of attacks by wild elephants while traveling through the streets and main roads in the NCP, as elephant threats are ubiquitous. This study aims to identify the environmental and human physiological factors contributing to incidents of human deaths caused by elephant attacks, and to predict the incidents of human deaths using some machine learning techniques, including Classification Trees (CT), Support Vector Machines (SVM), and Probabilistic Neural Networks (PNN), in the NCP. The department of Wildlife Conservation (DWC) and the department of Meteorology were contacted to collect data. On the other hand, the agricultural calendar was utilized to obtain information about the paddy harvest season. After identifying the key influential factors for human death incidences, the Chi-square test and Fisher's Exact test were employed to check the association between the response variable, incidents of human deaths and influential factors. Human gender, age group, rainfall season, paddy harvest season, and time period for encountering elephants were identified as significant factors for the response variable, incidents of human death. Overall Classification Accuracy (OCA) and Balanced Accuracy (BA) were used to measure the performance of each model, in which 80% of the data were randomly selected for the model-building process, and the remaining 20% were used to check model accuracy. Even though all the models demonstrated more than 85% OCA and BA, PNN exhibited the highest OCA of 88.57% and a BA of 88.39%, making it the better model to predict the incidents of human death due to elephant attacks. The final model and findings of this study can be used to ensure the safety of human lives in different incidents, as it reduces uncertainty of life. © 2025 IE

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Position-Specific Batting Role Assignment and Player Ranking Using Ensemble Learning 6

Position-Specific Batting Role Assignment and Player Ranking...

引用

6th IEEE International Conference on Artificial Intelligence in Engineering and Technology, IICAIET 2024

作者： Rajapaksha, R.A.R.S. Kumarika, B.M.T. University of Kelaniya Department of Statistics and Computer Science Sri Lanka

ISBN: (纸本)9798350389692

Cricket is a popular game around the world that includes three major formats. The T20 format stands out as the fastest format in cricket. A team's batting order significantly affects the outcome of the match. Every player must play his role according to the batting position. Hence assigning roles to batsmen based on their batting positions are crucial. Previous studies were performed on win prediction, team selection, player performance evaluation, and classification of players using machine learning. However, there are no prior studies to classify batsmen according to batting positions and rank players in each category. This study aims to classify players based on their roles in the batting order as top-order, middle-order, lower-order, or tail-ender batsman in T20 format and rank players in each category using classification probabilities. The data for the study was collected from *** website. For this study, 347 players who belong to a country that has Test membership were used. Initially, Classification was performed using machine learning models such as Naïve Bayes, RF (Random Forest), Decision Tree, SVM (Support Vector Machine), and KNN (K-Nearest Neighbors). Then, ensemble learning was used, and the best performance was observed in the stacking model with seven base models. That model reached a training accuracy of 97.69% and a testing accuracy of 95.40%. Additionally, it achieved a precision of 95.16% and a recall of 95.35%. The findings of this study show that ensemble models can effectively classify batsmen according to the batting positions and ranking players in each category. This classification and ranking are important to selectors, coaches, captains, and team management when making team selections, forming batting lineups, and performing tasks such as player replacement. The future work of this study aims to improve timely predictions using scheduled online training. © 2024 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Universally Instance-Optimal Mechanisms for Private Statistical Estimation 37

Universally Instance-Optimal Mechanisms for Private Statisti...

引用

37th Annual Conference on Learning Theory, COLT 2024

作者： Asi, Hilal Duchi, John C. Haque, Saminul Li, Zewei Ruan, Feng Apple United States Departments of Statistics and Electrical Engineering Stanford University United States Department of Computer Science Stanford University United States Department of Statistics and Data Science Northwestern University United States

We consider the problem of instance-optimal statistical estimation under the constraint of differential privacy where mechanisms must adapt to the difficulty of the input dataset. We prove a new instance specific lower bound using a new divergence and show it characterizes the local minimax optimal rates for private statistical estimation. We propose two new mechanisms that are universally instance-optimal for general estimation problems up to logarithmic factors. Our first mechanism, the total variation mechanism, builds on the exponential mechanism with stable approximations of the total variation distance, and is universally instance-optimal in the high privacy regime Ε ≤ 1/√n. Our second mechanism, the T-mechanism, is based on the T-estimator framework (Birgé, 2006) using the clipped log likelihood ratio as a stable test: it attains instance-optimal rates for any Ε ≤ 1 up to logarithmic factors. Finally, we study the implications of our results to robust statistical estimation, and show that our algorithms are universally optimal for this problem, characterizing the optimal minimax rates for robust statistical estimation. © 2024 H. Asi, J.C. Duchi, S. Haque, Z. Li & F. Ruan.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Semënov Arithmetic, Affine VASS, and String Constraints 41

Semënov Arithmetic, Affine VASS, and String Constraints

引用

41st International Symposium on Theoretical Aspects of computer science, STACS 2024

作者： Draghici, Andrei Haase, Christoph Manea, Florin Department of Computer Science University of Oxford United Kingdom Computer Science Department Göttingen University Campus-Institut Data Science Germany

ISBN: (纸本)9783959773119

We study extensions of Semënov arithmetic, the first-order theory of the structure (N,+,2x). It is well-known that this theory becomes undecidable when extended with regular predicates over tuples of number strings, such as the Büchi V2-predicate. We therefore restrict ourselves to the existential theory of Semënov arithmetic and show that this theory is decidable in EXPSPACE when extended with arbitrary regular predicates over tuples of number strings. Our approach relies on a reduction to the language emptiness problem for a restricted class of affine vector addition systems with states, which we show decidable in EXPSPACE. As an application of our result, we settle an open problem from the literature and show decidability of a class of string constraints involving length constraints. © Andrei Draghici, Christoph Haase, and Florin Manea;licensed under Creative Commons License CC-BY 4.0.

关键词： Computability and decidability

来源：评论

学校读者我要写书评

暂无评论

Evaluating Explanation Methods for Vision-and-Language Navigation 26

Evaluating Explanation Methods for Vision-and-Language Navig...

引用

26th European Conference on Artificial Intelligence, ECAI 2023

作者： Chen, Guanqi Yang, Lei Chen, Guanhua Pan, Jia Department of Computer Science The University of Hong Kong Hong Kong Hong Kong Department of Statistics and Data Science Southern University of Science and Technology China

ISBN: (纸本)9781643684369

The ability to navigate robots with natural language instructions in an unknown environment is a crucial step for achieving embodied artificial intelligence (AI). With the improving performance of deep neural models proposed in the field of vision-and-language navigation (VLN), it is equally interesting to know what information the models utilize for their decision-making in the navigation tasks. To understand the inner workings of deep neural models, various explanation methods have been developed for promoting explainable AI (XAI). But they are mostly applied to deep neural models for image or text classification tasks and little work has been done in explaining deep neural models for VLN tasks. In this paper, we address these problems by building quantitative benchmarks to evaluate explanation methods for VLN models in terms of faithfulness. We propose a new erasure-based evaluation pipeline to measure the step-wise textual explanation in the sequential decision-making setting. We evaluate several explanation methods for two representative VLN models on two popular VLN datasets and reveal valuable findings through our experiments. © 2023 The Authors.

关键词： Navigation

来源：评论

学校读者我要写书评

暂无评论

Composite Quantile Estimation for Kink Model with Longitudinal data

引用

Acta Mathematica Sinica,English Series 2023年第3期39卷 412-438页

作者： Chuang WAN Wei ZHONG Ying FANG School of Statistics and Data Science Nankai UniversityTianjin 300071P.R.China MOE Key Lab of Econometrics WISEDepartment of Statistics and Data ScienceSchool of EconomicsXiamen UniversityXiamen 361000P.R.China

Kink model is developed to analyze the data where the regression function is two-stage piecewise linear with respect to the threshold covariate but continuous at an unknown kink *** quantile regression for longitudinal data,kink point where the kink effect happens is often assumed to be heterogeneous across different ***,the kink point tends to be the same across different quantiles,especially in a region of neighboring quantile *** such homogeneity information could increase the estimation efficiency of the common kink *** this paper,we propose a composite quantile estimation approach for the common kink point by combining information from multiple neighboring *** normality of the proposed estimator is *** addition,we also develop a sup-likelihood-ratio test to check the existence of the kink effect at a given quantile level.A test-inversion confidence interval for the common kink point is also developed based on the quantile rank score *** simulation studies show that the proposed composite kink estimator is more efficient than the single quantile regression *** also illustrate the proposed method via an application to a longitudinal data set on blood pressure and body mass index.

关键词： Asymptotical normality composite quantile estimation estimation efficiency kink design model longitudinal data

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：