检索结果-内蒙古大学图书馆

preference learning for efficient bundle selection in horizontal transport collaborations

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH 2025年第3期324卷 953-968页

作者： Elting, Steffen Ehmke, Jan Fabian Gansterer, Margaretha Univ Vienna Dept Business Decis & Analyt Kolingasse 14-16 A-1090 Vienna Austria Univ Klagenfurt Dept Prod Management & Logist Univ Str 65-67 A-9020 Klagenfurt Worthersee Austria

To improve routing efficiency, transport service providers can enter a horizontal transport collaboration that uses a combinatorial auction to reallocate delivery orders. To find the optimal allocation, the carriers have to report bids for all possible combinations of available delivery orders. As this number grows exponentially with the number of orders to be reallocated, they are faced with an enormous computational challenge. To lift this burden, the auctioneer may offer only a limited set of order combinations. However, selecting this limited set is itself a stochastic combinatorial optimization problem known as the Bundle Selection Problem. In contrast to previous one-shot approaches to solve this problem, in this paper, a partial preference learning scheme is applied that iteratively queries carriers' valuations, uses their responses to train preference models and then uses these fitted models to estimate valuations for new combinations of orders. This work investigates different ways to realize such a concept and analyzes their respective improvement in collaboration gains. The results indicate that the suggested algorithm can yield travel time savings of up to 20% higher than those achieved by a random benchmark and up to 10% higher than those of a literature benchmark if at least 40 query-response pairs are considered.

关键词： Routing preference learning Collaborative transportation planning Combinatorial auction Vehicle routing problem with time windows

来源：评论

学校读者我要写书评

暂无评论

preference learning based deep reinforcement learning for flexible job shop scheduling problem

引用

COMPLEX & INTELLIGENT SYSTEMS 2025年第2期11卷 1-23页

作者： Liu, Xinning Han, Li Kang, Ling Liu, Jiannan Miao, Huadong Dalian Neusoft Univ Informat Sch Comp & Software Dalian 116023 Liaoning Peoples R China Dalian Neusoft Univ Informat Neusoft Res Inst Dalian 116023 Liaoning Peoples R China SNOW China Beijing Co Ltd Dalian Branch Dalian 116023 Liaoning Peoples R China

The flexible job shop scheduling problem (FJSP) holds significant importance in both theoretical research and practical applications. Given the complexity and diversity of FJSP, improving the generalization and quality of scheduling methods has become a hot topic of interest in both industry and academia. To address this, this paper proposes a preference-Based Mask-PPO (PBMP) algorithm, which leverages the strengths of preference learning and invalid action masking to optimize FJSP solutions. First, a reward predictor based on preference learning is designed to model reward prediction by comparing random fragments, eliminating the need for complex reward function design. Second, a novel intelligent switching mechanism is introduced, where proximal policy optimization (PPO) is employed to enhance exploration during sampling, and masked proximal policy optimization (Mask-PPO) refines the action space during training, significantly improving efficiency and solution quality. Furthermore, the Pearson correlation coefficient (PCC) is used to evaluate the performance of the preference model. Finally, comparative experiments on FJSP benchmark instances of varying sizes demonstrate that PBMP outperforms traditional scheduling strategies such as dispatching rules, OR-Tools, and other deep reinforcement learning (DRL) algorithms, achieving superior scheduling policies and faster convergence. Even with increasing instance sizes, preference learning proves to be an effective reward mechanism in reinforcement learning for FJSP. The ablation study further highlights the advantages of each key component in the PBMP algorithm across performance metrics.

关键词： Flexible job shop scheduling problem preference learning Proximal policy optimization Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

preference learning Based on Dynamic Dual-Cognition for Group Recommendation 35th

Preference Learning Based on Dynamic Dual-Cognition for Grou...

引用

35th Australasian Database Conference

作者： Wang, Chunlong Kou, Yue Shen, Derong Zhou, Xiangmin Nie, Tiezheng Yu, Ge Li, Dong Northeastern Univ Sch Comp Sci & Engn Shenyang 110004 Liaoning Peoples R China RMIT Univ Sch Comp Technol Melbourne Vic 3000 Australia Liaoning Univ Coll Informat Shenyang 110036 Liaoning Peoples R China

ISBN: (纸本)9789819612413;9789819612420

Group recommendation aims to suggest items that cater to the preferences of all members within the group. However, existing models often overlook the dynamic cognitive changes of group members, leading to inaccuracies. In this paper, we propose a novel preference learning framework based on Dynamic Dual-cognition for Group Recommendation (PL-DDGR), which aims to enhance group recommendation accuracy. Specifically, we first propose a preference learning approach based on graduality cognition, which can better understand and predict the subtle yet continuous shifts in member preferences. Then we propose a preference learning approach based on conformity cognition, which can capture the evolving nature of member conformity. We also propose a self-supervised multi-task joint training mechanism to optimize the learning of both graduality and conformity cognition simultaneously. The experiments demonstrate the effectiveness and superiority of our proposed framework.

关键词： Group Recommendation preference learning Graduality Cognition Conformity Cognition

来源：评论

学校读者我要写书评

暂无评论

Data-driven preference learning methods for sorting problems with multiple temporal criteria

引用

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH 2025年第3期323卷 918-937页

作者： Li, Yijun Guo, Mengzhuo Kadzinski, Milosz Zhang, Qingpeng Xu, Chenxi City Univ Hong Kong Sch Data Sci 83 Tat Chee Ave Kowloon Tong Hong Kong Peoples R China Sichuan Univ Business Sch 24 South Sect 1Yihuan Rd Chengdu 610065 Peoples R China Poznan Univ Tech Inst Comp Sci Piotrowo 2 PL-60965 Poznan Poland Univ Hong Kong Musketeers Fdn Inst Data Sci Pokfulam Hong Kong Peoples R China Univ Hong Kong LKS Fac Med Pokfulam Hong Kong Peoples R China

We present novel preference learning approaches for sorting problems with multiple temporal criteria. They leverage an additive value function as the basic preference model, adapted for accommodating time series data. Given assignment examples concerning reference alternatives, we learn such a model using convex quadratic programming. It is characterized by fixed-time discount factors and operates within a regularization framework. This approach enables the consideration of temporal interdependencies between timestamps while mitigating the risk of overfitting. To enhance scalability and accommodate learnable time discount factors, we introduce a novel monotonic Recurrent Neural Network (mRNN). It captures the evolving dynamics of preferences over time, while upholding critical properties inherent to multiple criteria sorting problems. These include criteria monotonicity, preference independence, and the natural ordering of classes. The proposed mRNN can describe the preference dynamics by depicting piecewise linear marginal value functions and personalized time discount factors along with time. Thus, it effectively combines the interpretability of traditional sorting methods with the predictive potential offered by deep preference learning models. We comprehensively assess the proposed models on synthetic data scenarios and a real-case study centered on classifying valuable users within a mobile gaming app based on their historical in-game behavioral sequences. Empirical findings underscore the notable performance improvements achieved by the proposed models when compared to a spectrum of baseline methods, spanning machine learning, deep learning, and conventional multiple criteria sorting approaches.

关键词： Multiple criteria analysis preference learning Sorting Temporal criteria Deep learning Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Determining investment allocation strategies to improve consumer satisfaction based on a preference learning model

引用

JOURNAL OF RETAILING AND CONSUMER SERVICES 2025年 82卷

作者： Wu, Xingli Liao, Huchang Sichuan Univ Business Sch Chengdu 610064 Peoples R China

Mining product attribute performance, importance, and their (a)symmetric impacts on consumer satisfaction from online reviews is crucial for enterprises to formulate real-time investment allocation strategies for product improvement. While existing studies have employed machine learning, regression, and correlation analysis to explore these complex relationships, they face the challenge of balancing prediction accuracy with interpretability. This paper proposes an asymmetric importance-performance analysis model based on preference learning with online reviews. It devises an asymmetric value function incorporating unknown preference parameters to elucidate (a)symmetric impacts of attribute performance on overall consumer satisfaction. The process of learning preference parameters is implemented by mathematical programming with a simulation experiment. Attributes are classified into eight categories according to their performance and importance, each corresponding to an improvement strategy. An optimization model is constructed to develop investment allocation strategies for attribute improvement, aiming at maximizing consumer satisfaction within established financial constraints. A hotel-focused case study showcases the approach, and simulations validate the robustness of the proposed model.

关键词： preference learning Consumer satisfaction analysis Product improvement Online reviews Importance-performance analysis

来源：评论

学校读者我要写书评

暂无评论

Crowdsourcing Task Assignment with Category and Mobile Combined preference learning 22nd

Crowdsourcing Task Assignment with Category and Mobile Combi...

引用

22nd International Conference on Service Oriented Computing

作者： Ma, Yue Gao, Xiaofeng Chen, Guihai Shanghai Jiao Tong Univ Dept Comp Sci & Engn MoE Key Lab Artificial Intelligence Shanghai 200240 Peoples R China

ISBN: (纸本)9789819608041;9789819608058

The rapid growth of mobile networks and widespread mobile device usage has brought mobile crowdsourcing (MCS) for location-sensitive services into focus, particularly for spatial task assignments. However, current research often overlooks the combined impact of users' category and mobile preferences, resulting in suboptimal task assignments. Additionally, the inherent heterogeneity among users is frequently ignored, failing to represent the true dynamics of the MCS ecosystem. To address these gaps, we propose a comprehensive framework, Task Assignment with User preference and Heterogeneity, consisting of two key components: the Category and Mobile Combined preference (CAMP) model and the preference-Aware Task Assignment mechanism. The CAMP model predicts users' combined category and mobile preferences using attention mechanisms to extract insights from sparse historical data. In parallel, the preference-Aware Task Assignment mechanism introduces three novel algorithms that account for user preference and capacity heterogeneity. Extensive experiments on real-world datasets demonstrate the effectiveness and efficiency of the proposed methods.

关键词： Task assignment Mobile crowdsourcing preference learning Attention mechanism

来源：评论

学校读者我要写书评

暂无评论

VisQ2SQL: Towards SQL-Driven Data Visualization via LLMs-Grounded preference learning

VisQ2SQL: Towards SQL-Driven Data Visualization via LLMs-Gro...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Shi, Shengze Ren, Tao Hu, Jun State Key Laboratory of Intelligent Game Institute of Software Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9798350368741

Text-to-Visualization (Text2Vis) aims to democratize data insights for non-expert users by transforming natural language query (NLQ) into visualization specification (VS). In view of the high dependence of rule-based methods on predefined VS templates and poor NLQ understanding ability of small data-driven methods, recent works resort to leveraging pre-trained LLMs to perform NLQ-understanding and VS-generating in Text2Vis tasks via prompt-guided in-context learning. However, existing LLM-based methods still fall short of satisfactory end-to-end Text2Vis performances primarily owing to the limited ability of pre-trained LLMs in directly retrieving and operating NLQ-intended tabular data. Inspired by the SQL generating ability born with latest LLMs, this paper proposes the idea of harnessing LLMs for SQL-driven visualization data retrieval and operation. Nonetheless, there remains a nonneglectable gap between data visualization queries in Text2Vis tasks and SQL retrieving queries in LLM corpus. To fill the gap, this paper proposes a Visualization Query to SQL (VisQ2SQL) framework to obtain NLQ-intended data, primarily by fine-tuning LLMs through preference learning data-retrieval SQLs induced from VS and those generated by LLMs. We conduct extensive experiments to demonstrate the superiority of VisQ2SQL over SOTA methods, and various ablation studies to verify the efficacy of VisQ2SQL. © 2025 IEEE.

关键词： Large Language Model preference learning Text to Visualization

来源：评论

学校读者我要写书评

暂无评论

A data-driven preference learning approach for multi-objective vehicle routing problems in last-mile delivery

引用

TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES 2025年 174卷

作者： Nourmohammadi, Zahra Hu, Bohan Rey, David Saberi, Meead Univ New South Wales Sch Civil & Environm Engn Res Ctr Integrated Transport Innovat Sydney 2052 Australia Univ Cote Azur SKEMA Business Sch Sophia Antipolis France

Last-mile delivery service providers and drivers often choose routes deviating from the shortest distance, influenced by personal preferences and various business-level factors. This study introduces an innovative data-driven optimization approach for learning preferences of decision- makers (DMs) in multi-objective vehicle routing problems. Utilizing real-world historical data from a last-mile delivery logistics platform, we develop a machine-learning framework to learn DMs' preferences in designing delivery routes. To design our approach we focus on a multi- objective capacitated vehicle routing problem with time windows and develop an integrated framework that combines supervised learning models, sampling techniques, and optimization methods to determine preference weights for objective functions based on selected features. We conduct extensive numerical experiments to test the proposed data-driven optimization approach. Our findings suggest that analyzing historical planned and actual routes reveals DMs' preferences, such as prioritizing workload balance and minimizing fleet usage over travel distance alone. Furthermore, this study offers insights into key factors shaping last-mile delivery logistics, including workload distribution and deviations from pre-planned routes, enabling more informed and human-centered decision-making in logistics optimization.

关键词： Multi-objective vehicle routing preference learning Data-driven optimization Machine learning

来源：评论

学校读者我要写书评

暂无评论

Modeling biodiesel properties by preference learning: Case study of cetane number

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 144卷

作者： Diez-Valbuena, Guillermo Garcia-Tuero, Alejandro Rodriguez, Eduardo Battez, Antolin Hernandez Diez, Jorge Univ Oviedo Dept Construct & Mfg Engn Pedro Puig Adam S-N Gijon 33203 Spain Univ Oviedo Artificial Intelligence Ctr Campus Gijon Gijon 33203 Spain

The properties and performance of biodiesels depend on the composition of the fatty acid methyl esters (FAMEs) present. Obtaining new biodiesels is neither a cheap nor a simple task. Therefore, it is important to model their properties in advance in order to know what composition would be needed to fulfill a given requirement. This paper presents a multitask approach that models the behavior of biodiesels using preference learning and also obtains representations of those biodiesels in a feature space by means of an autoencoder. The performance of the system in different configurations is also analyzed, showing its robustness. A case study was conducted to analyze the cetane number, which is an important property in biodiesel. Correlations between the cetane number and the distribution of FAMEs have been identified. These correlations are more clearly observed when the FAMEs are grouped according to common characteristics, such as the percentage of saturated fatty acids (SFA), monounsaturated fatty acids (MUFA), and polyunsaturated fatty acids (PUFA) in the biodiesel, or the average number of double bonds (DB). Based on these groupings, it can be observed that the cetane number decreases with increasing unsaturation. Additionally, it was determined that the cetane number value has a minimal or no correlation with the average molecular weight (Mw) of the biodiesel.

关键词： preference learning Autoencoder Multitask Siamese network Biodiesel Cetane number

来源：评论

学校读者我要写书评

暂无评论

Utility-Driven Minimum Cost Consensus Model With preference learning and Fairness Concerns

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2025年

作者： Zhang, Bowen Huang, Wenjun Dong, Yucheng Pedrycz, Witold Cent South Univ Business Sch Changsha 410083 Peoples R China Sichuan Univ Business Sch Chengdu 610065 Peoples R China Univ Alberta Dept Elect & Comp Engn Edmonton AB T6G 2R3 Canada Macau Univ Sci & Technol Inst Syst Engn Macau Peoples R China Polish Acad Sci Syst Res Inst PL-00901 Warsaw Poland Istinye Univ Res Ctr Performance & Prod Anal TR-34396 Istanbul Turkiye

Nowadays the emergence of digital technology and artificial intelligence brings a lot of extra information and convenience for group decision activities, such as rich historical decision data, mobile decision platform, artificial interactive device and others. However, it also results in high complexity and uncertainty of the decision environments, modeling and analysis, and the preference of decision maker is crucial to the design of group decision and consensus mechanism. Meanwhile, the information transparency results in more attention being paid to fairness concerns in group decision making. Based on the historical group decision data, this study proposes a novel minimum cost consensus model with preference learning and fairness concerns (MCCM-PLFC), which characterizes the modification motivation of decision makers in a utility-driven way. A preference learning framework is first proposed to determine the parameters of utility functions of decision makers, where fairness concerns are modeled based on the return bias among the decision makers. Then, in a Stackelberg game consensus framework, the MCCM-PLFC is constructed to minimize the consensus cost of the moderator with consideration of the utility-driven behavior of decision makers. In addition, a best-response update algorithm is designed to solve the equilibrium solution among the decision makers and an adaptive differential evolution algorithm is used to solve the MCCM-PLFC. Finally, a throughout of experimental study is performed to quantify the validity of the proposed consensus model.

关键词： Costs Games Adaptation models Data models Consensus protocol Analytical models Predictive models Optimization models Nash equilibrium Hands Consensus cost consensus model fairness concerns preference learning Stackelberg game

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：