Information Processing & Management最新文献

筛选
英文 中文
The joint extraction of fact-condition statement and super relation in scientific text with table filling method 用表格填充法联合提取科学文本中的事实条件陈述和超级关系
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-10-01 DOI: 10.1016/j.ipm.2024.103906
Qizhi Chen , Hong Yao , Diange Zhou
{"title":"The joint extraction of fact-condition statement and super relation in scientific text with table filling method","authors":"Qizhi Chen ,&nbsp;Hong Yao ,&nbsp;Diange Zhou","doi":"10.1016/j.ipm.2024.103906","DOIUrl":"10.1016/j.ipm.2024.103906","url":null,"abstract":"<div><div>The fact-condition statements are of great significance in scientific text, via which the natural phenomenon and its precondition are detailly recorded. In previous study, the extraction of fact-condition statement and their relation (super relation) from scientific text is designed as a pipeline that the fact-condition statement and super relation are extracted successively, which leads to the error propagation and lowers the accuracy. To solve this problem, the table filling method is firstly adopted for joint extraction of fact-condition statement and super relation, and the Biaffine Convolution Neural Network model (BCNN) is proposed to complete the task. In the BCNN, the pretrained language model and Biaffine Neural Network work as the encoder, while the Convolution Neural Network is added into the model as the decoder that enhances the local semantic information. Benefiting from the local semantic enhancement, the BCNN achieves the best F1 score with different pretrained language models in comparison with other baselines. Its F1 scores in GeothCF (geological text) reach 73.17% and 71.04% with BERT and SciBERT as pretrained language model, respectively. Moreover, the local semantic enhancement also increases its training efficiency, via which the tags’ distribution can be more easily learned by the model. Besides, the BCNN trained with GeothCF also exhibits the best performance in BioCF (biomedical text), which indicates that it can be widely applied for the information extraction in all scientific domains. Finally, the geological fact-condition knowledge graph is built with BCNN, showing a new pipeline for construction of scientific fact-condition knowledge graph.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103906"},"PeriodicalIF":7.4,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142417752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Crowdsourced auction-based framework for time-critical and budget-constrained last mile delivery 基于众包拍卖的时间紧迫、预算有限的最后一英里交付框架
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-30 DOI: 10.1016/j.ipm.2024.103888
Esraa Odeh , Shakti Singh , Rabeb Mizouni , Hadi Otrok
{"title":"Crowdsourced auction-based framework for time-critical and budget-constrained last mile delivery","authors":"Esraa Odeh ,&nbsp;Shakti Singh ,&nbsp;Rabeb Mizouni ,&nbsp;Hadi Otrok","doi":"10.1016/j.ipm.2024.103888","DOIUrl":"10.1016/j.ipm.2024.103888","url":null,"abstract":"<div><div>This work addresses the problem of Last Mile Delivery (LMD) under time-critical and budget-constrained environments. Given the rapid growth of e-commerce worldwide, LMD has become a primary bottleneck to the efficiency of delivery services due to several factors, including travelling distance, service cost, and delivery time. Existing works mainly target optimizing travelled distance and maximizing gained profit; however, they do not consider time-critical and budget-limited tasks. The deployment of UAVs and the development of crowdsourcing platforms have provided a range of solutions to advance performance in LMD frameworks, as they offer many crowdworkers at varying locations ready to perform tasks instead of having a single point of departure. This work proposes a Hybrid, Crowdsourced, Auction-based LMD (HCA-LMD) framework with a dynamic allocation mechanism for optimized delivery of time-sensitive and budget-limited tasks. The proposed framework allocates tasks to workers as soon as they are submitted, given their urgency level and dropoff location, while considering the price, rating, and location of available workers. This work was compared against two benchmarks to assess the framework’s performance in dynamic environments in terms of on-time deliveries, average delay, and profit. Extensive simulation results showed an outstanding performance of the proposed state-of-the-art LMD framework by accomplishing almost 92% on-time deliveries under varying time- and budget-constrained scenarios, outperforming the first benchmark in the on-time allocation rate by fulfiling an additional 24% of the tasks the benchmark failed, with around 50% drop in average delay time and up to x5.8 gained profit when compared against the second benchmark.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103888"},"PeriodicalIF":7.4,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
An interpretable polytomous cognitive diagnosis framework for predicting examinee performance 用于预测考生成绩的可解释多项式认知诊断框架
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-29 DOI: 10.1016/j.ipm.2024.103913
Xiaoyu Li , Shaoyang Guo , Jin Wu , Chanjin Zheng
{"title":"An interpretable polytomous cognitive diagnosis framework for predicting examinee performance","authors":"Xiaoyu Li ,&nbsp;Shaoyang Guo ,&nbsp;Jin Wu ,&nbsp;Chanjin Zheng","doi":"10.1016/j.ipm.2024.103913","DOIUrl":"10.1016/j.ipm.2024.103913","url":null,"abstract":"<div><div>As a fundamental task of intelligent education, deep learning-based cognitive diagnostic models (CDMs) have been introduced to effectively model dichotomous testing data. However, it remains a challenge to model the polytomous data within the deep-learning framework. This paper proposed a novel <strong>P</strong>olytomous <strong>C</strong>ognitive <strong>D</strong>iagnosis <strong>F</strong>ramework (PCDF), which employs <strong>C</strong>umulative <strong>C</strong>ategory <strong>R</strong>esponse <strong>F</strong>unction (CCRF) theory to partition and consolidate data, thereby enabling existing cognitive diagnostic models to seamlessly analyze graded response data. By combining the proposed PCDF with IRT, MIRT, NCDM, KaNCD, and ICDM, extensive experiments were complemented by data re-encoding techniques on the four real-world graded scoring datasets, along with baseline methods such as linear-split, one-vs-all, and random. The results suggest that when combined with existing CDMs, PCDF outperforms the baseline models in terms of prediction. Additionally, we showcase the interpretability of examinee ability and item parameters through the utilization of PCDF.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103913"},"PeriodicalIF":7.4,"publicationDate":"2024-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A theoretical framework for human-centered intelligent information services: A systematic review 以人为本的智能信息服务理论框架:系统回顾
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-29 DOI: 10.1016/j.ipm.2024.103891
Qiao Li, Yuelin Li, Shuhan Zhang, Xin Zhou, Zhengyuan Pan
{"title":"A theoretical framework for human-centered intelligent information services: A systematic review","authors":"Qiao Li,&nbsp;Yuelin Li,&nbsp;Shuhan Zhang,&nbsp;Xin Zhou,&nbsp;Zhengyuan Pan","doi":"10.1016/j.ipm.2024.103891","DOIUrl":"10.1016/j.ipm.2024.103891","url":null,"abstract":"<div><div>Intelligent Information Services (IIS) employ Artificial Intelligence (AI)-based systems to provide information that matches the user's needs in diverse and evolving environments. Acknowledging the importance of users in AI-empowered IIS success, a growing number of researchers are investigating AI-empowered IIS from a user-centric perspective, establishing the foundation for a new research domain called “Human-Centered Intelligent Information Services” (HCIIS). Nonetheless, a review of user studies in AI-empowered IIS is still lacking, impeding the development of a clear definition and research framework for the HCIIS field. To fill this gap, this study conducts a systematic review of 116 user studies in AI-empowered IIS. Results reveal two primary research themes in user studies in AI-empowered IIS: human-IIS interaction (including user experience, system quality, user attitude, intention and behavior, information quality, and individual task performance) and IIS ethics (e.g., explainability and interpretability, privacy and safety, and inclusivity). Analyzing research gaps within these topics, this study formulates an HCIIS research framework consisting of three interconnected elements: human values and needs, environment, and service. The interconnections between each pair of elements identify three key research domains in HCIIS: interaction, ethics, and evolution. Interaction pertains to the facilitation of human-IIS interaction to meet human needs, encompassing topics including human-centered theory, evaluation, and the design of AI-empowered IIS interaction. Ethics emphasize ensuring AI-empowered IIS alignment with human values and norms within specific environments, covering topics like general and context-specific AI-empowered IIS ethical principles, risk assessment, and governance strategies. Evolution focuses on addressing the fulfillment of human needs in diverse and dynamic environments by continually evolving intelligence, involving the enhancement of AI-empowered IIS environmental sensitivity and adaptability within an intelligent ecosystem driven by technology integration. Central to HCIIS is co-creation, situated at the intersection of interaction, evolution, and ethics, emphasizing collaborative information creation between IIS and humans through hybrid intelligence. In conclusion, HCIIS is defined as a field centered on information co-creation between IIS and humans, distinguishing it from IIS, which focuses on providing information to humans.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103891"},"PeriodicalIF":7.4,"publicationDate":"2024-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
DualFLAT: Dual Flat-Lattice Transformer for domain-specific Chinese named entity recognition DualFLAT:用于特定领域中文命名实体识别的双平面-网格变换器
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-28 DOI: 10.1016/j.ipm.2024.103902
Yinlong Xiao , Zongcheng Ji , Jianqiang Li , Qing Zhu
{"title":"DualFLAT: Dual Flat-Lattice Transformer for domain-specific Chinese named entity recognition","authors":"Yinlong Xiao ,&nbsp;Zongcheng Ji ,&nbsp;Jianqiang Li ,&nbsp;Qing Zhu","doi":"10.1016/j.ipm.2024.103902","DOIUrl":"10.1016/j.ipm.2024.103902","url":null,"abstract":"<div><div>Recently, lexicon-enhanced methods for Chinese Named Entity Recognition (NER) have achieved great success which requires a high-quality lexicon. However, for the domain-specific Chinese NER, it is challenging to obtain such a high-quality lexicon due to the different distribution between the general lexicon and domain-specific data, and the high construction cost of the domain lexicon. To address these challenges, we introduce dual-source lexicons (<em>i.e.,</em> a general lexicon and a domain lexicon) to acquire enriched lexical knowledge. Considering that the general lexicon often contains more noise compared to its domain counterparts, we further propose a dual-stream model, Dual Flat-LAttice Transformer (DualFLAT), designed to mitigate the impact of noise originating from the general lexicon while comprehensively harnessing the knowledge contained within the dual-source lexicons. Experimental results on three public domain-specific Chinese NER datasets (<em>i.e.,</em> News, Novel and E-commerce) demonstrate that our method consistently outperforms the single-source lexicon-enhanced approaches, achieving state-of-the-art results. Specifically, our proposed DualFLAT model consistently outperforms the baseline FLAT, with an increase of up to 1.52%, 4.84% and 1.34% in F1 score for the News, Novel and E-commerce datasets, respectively.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103902"},"PeriodicalIF":7.4,"publicationDate":"2024-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Gauging, enriching and applying geography knowledge in Pre-trained Language Models 衡量、丰富和应用预训练语言模型中的地理知识
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-27 DOI: 10.1016/j.ipm.2024.103892
Nitin Ramrakhiyani , Vasudeva Varma , Girish Keshav Palshikar , Sachin Pawar
{"title":"Gauging, enriching and applying geography knowledge in Pre-trained Language Models","authors":"Nitin Ramrakhiyani ,&nbsp;Vasudeva Varma ,&nbsp;Girish Keshav Palshikar ,&nbsp;Sachin Pawar","doi":"10.1016/j.ipm.2024.103892","DOIUrl":"10.1016/j.ipm.2024.103892","url":null,"abstract":"<div><div>To employ Pre-trained Language Models (PLMs) as knowledge containers in niche domains it is important to gauge the knowledge of these PLMs about facts in these domains. It is also an important pre-requisite to know how much enrichment effort is required to make them better. As part of this work, we aim to gauge and enrich small PLMs for knowledge of world geography. Firstly, we develop a moderately sized dataset of masked sentences covering 24 different fact types about world geography to estimate knowledge of PLMs on these facts. We hypothesize that for this niche domain, smaller PLMs may not be well equipped. Secondly, we enrich PLMs with this knowledge through fine-tuning and check if the knowledge in the dataset is infused sufficiently. We further hypothesize that linguistic variability in the manual templates used to embed the knowledge in masked sentences does not affect the knowledge infusion. Finally, we demonstrate the application of PLMs to tourism blog search and Wikidata KB augmentation. In both applications, we aim at showing the effectiveness of using PLMs to achieve competitive performance.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103892"},"PeriodicalIF":7.4,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142326351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
DST: Continual event prediction by decomposing and synergizing the task commonality and specificity DST:通过分解和协同任务的共性和特殊性来进行连续事件预测
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-26 DOI: 10.1016/j.ipm.2024.103899
Yuxin Zhang , Songlin Zhai , Yongrui Chen , Shenyu Zhang , Sheng Bi , Yuan Meng , Guilin Qi
{"title":"DST: Continual event prediction by decomposing and synergizing the task commonality and specificity","authors":"Yuxin Zhang ,&nbsp;Songlin Zhai ,&nbsp;Yongrui Chen ,&nbsp;Shenyu Zhang ,&nbsp;Sheng Bi ,&nbsp;Yuan Meng ,&nbsp;Guilin Qi","doi":"10.1016/j.ipm.2024.103899","DOIUrl":"10.1016/j.ipm.2024.103899","url":null,"abstract":"<div><div>Event prediction aims to forecast future events by analyzing the inherent development patterns of historical events. A desirable event prediction system should learn new event knowledge, and adapt to new domains or tasks that arise in real-world application scenarios. However, continuous training can lead to catastrophic forgetting of the model. While existing continuous learning methods can retain characteristic knowledge from previous domains, they ignore potential shared knowledge in subsequent tasks. To tackle these challenges, we propose a novel event prediction method based on graph structural commonality and domain characteristic prompts, which not only avoids forgetting but also facilitates bi-directional knowledge transfer across domains. Specifically, we mitigate model forgetting by designing domain characteristic-oriented prompts in a continuous task stream with frozen the backbone pre-trained model. Building upon this, we further devise a commonality-based adaptive updating algorithm by harnessing a unique structural commonality prompt to inspire implicit common features across domains. Our experimental results on two public benchmark datasets for event prediction demonstrate the effectiveness of our proposed continuous learning event prediction method compared to state-of-the-art baselines. In tests conducted on the IED-Stream, DST’s ET-TA metric significantly improved by 5.6% over the current best baseline model, while the ET-MD metric, which reveals forgetting, decreased by 5.8%.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103899"},"PeriodicalIF":7.4,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
An adaptive confidence-based data revision framework for Document-level Relation Extraction 用于文档级关系提取的基于置信度的自适应数据修订框架
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-26 DOI: 10.1016/j.ipm.2024.103909
Chao Jiang , Jinzhi Liao , Xiang Zhao , Daojian Zeng , Jianhua Dai
{"title":"An adaptive confidence-based data revision framework for Document-level Relation Extraction","authors":"Chao Jiang ,&nbsp;Jinzhi Liao ,&nbsp;Xiang Zhao ,&nbsp;Daojian Zeng ,&nbsp;Jianhua Dai","doi":"10.1016/j.ipm.2024.103909","DOIUrl":"10.1016/j.ipm.2024.103909","url":null,"abstract":"<div><div>Noisy annotations have become a key issue limiting <strong>Doc</strong>ument-level <strong>R</strong>elation <strong>E</strong>xtraction <strong>(DocRE)</strong>. Previous research explored the problem through manual re-annotation. However, the handcrafted strategy is of low efficiency, incurs high human costs and cannot be generalized to large-scale datasets. To address the problem, we construct a confidence-based <strong>Re</strong>vision framework for <strong>D</strong>ocRE (<strong>ReD</strong>), aiming to achieve high-quality automatic data revision. Specifically, we first introduce a denoising training module to recognize relational facts and prevent noisy annotations. Second, a confidence-based data revision module is equipped to perform adaptive data revision for long-tail distributed relational facts. After the data revision, we design an iterative training module to create a virtuous cycle, which transforms the revised data into useful training data to support further revision. By capitalizing on ReD, we propose <strong>ReD-DocRED</strong>, which consists of 101,873 revised annotated documents from DocRED. ReD-DocRED has introduced 57.1% new relational facts, and concurrently, models trained on ReD-DocRED have achieved significant improvements in F1 scores, ranging from 6.35 to 16.55. The experimental results demonstrate that ReD can achieve high-quality data revision and, to some extent, replace manual labeling.<span><span><sup>1</sup></span></span></div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103909"},"PeriodicalIF":7.4,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Mitigating the negative impact of over-association for conversational query production 减轻过度关联对会话查询制作的负面影响
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-26 DOI: 10.1016/j.ipm.2024.103907
Ante Wang , Linfeng Song , Zijun Min , Ge Xu , Xiaoli Wang , Junfeng Yao , Jinsong Su
{"title":"Mitigating the negative impact of over-association for conversational query production","authors":"Ante Wang ,&nbsp;Linfeng Song ,&nbsp;Zijun Min ,&nbsp;Ge Xu ,&nbsp;Xiaoli Wang ,&nbsp;Junfeng Yao ,&nbsp;Jinsong Su","doi":"10.1016/j.ipm.2024.103907","DOIUrl":"10.1016/j.ipm.2024.103907","url":null,"abstract":"<div><div>Conversational query generation aims at producing search queries from dialogue histories, which are then used to retrieve relevant knowledge from a search engine to help knowledge-based dialogue systems. Trained to maximize the likelihood of gold queries, previous models suffer from the data hunger issue, and they tend to both drop important concepts from dialogue histories and generate irrelevant concepts at inference time. We attribute these issues to the <em>over-association</em> phenomenon where a large number of gold queries are indirectly related to the dialogue topics, because annotators may unconsciously perform reasoning with their background knowledge when generating these gold queries. We carefully analyze the negative effects of this phenomenon on pretrained Seq2seq query producers and then propose effective instance-level weighting strategies for training to mitigate these issues from multiple perspectives. Experiments on two benchmarks, Wizard-of-Internet and DuSinc, show that our strategies effectively alleviate the negative effects and lead to significant performance gains (2%<!--> <span><math><mo>∼</mo></math></span> <!--> <!-->5% across automatic metrics and human evaluation). Further analysis shows that our model selects better concepts from dialogue histories and is <em>10 times</em> more data efficient than the baseline.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103907"},"PeriodicalIF":7.4,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A comprehensive study on fidelity metrics for XAI 关于 XAI 真实度指标的综合研究
IF 7.4 1区 管理学
Information Processing & Management Pub Date : 2024-09-26 DOI: 10.1016/j.ipm.2024.103900
Miquel Miró-Nicolau, Antoni Jaume-i-Capó, Gabriel Moyà-Alcover
{"title":"A comprehensive study on fidelity metrics for XAI","authors":"Miquel Miró-Nicolau,&nbsp;Antoni Jaume-i-Capó,&nbsp;Gabriel Moyà-Alcover","doi":"10.1016/j.ipm.2024.103900","DOIUrl":"10.1016/j.ipm.2024.103900","url":null,"abstract":"<div><div>The use of eXplainable Artificial Intelligence (XAI) systems has introduced a set of challenges that need resolution. Herein, we focus on how to correctly select an XAI method, an open questions within the field. The inherent difficulty of this task is due to the lack of a ground truth. Several authors have proposed metrics to approximate the fidelity of different XAI methods. These metrics lack verification and have concerning disagreements. In this study, we proposed a novel methodology to verify fidelity metrics, using transparent models. These models allowed us to obtain explanations with perfect fidelity. Our proposal constitutes the first objective benchmark for these metrics, facilitating a comparison of existing proposals, and surpassing existing methods. We applied our benchmark to assess the existing fidelity metrics in two different experiments, each using public datasets comprising 52,000 images. The images from these datasets had a size a 128 by 128 pixels and were synthetic data that simplified the training process. We identified that two fidelity metrics, Faithfulness Estimate and Faithfulness Correlation, obtained the expected perfect results for linear models, showing their ability to approximate fidelity for this kind of methods. However, when present with non-linear models, as the ones most used in the state-of-the-art,all metric values, indicated a lack of fidelity, with the best one showing a 30% deviation from the expected values for perfect explanation. Our experimentation led us to conclude that the current fidelity metrics are not reliable enough to be used in real scenarios. From this finding, we deemed it necessary to development new metrics, to avoid the detected problems, and we recommend the usage of our proposal as a benchmark within the scientific community to address these limitations.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 1","pages":"Article 103900"},"PeriodicalIF":7.4,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信