본문

서브메뉴

Explaining Unintuitive Feature Importance Explanations
Explaining Unintuitive Feature Importance Explanations
Explaining Unintuitive Feature Importance Explanations

Detailed Information

자료유형  
 학위논문 서양
최종처리일시  
20260202103209
ISBN  
9798291558157
DDC  
020
저자명  
Qu, Jiaming.
서명/저자  
Explaining Unintuitive Feature Importance Explanations
발행사항  
[Sl] : The University of North Carolina at Chapel Hill, 2025
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2025
형태사항  
166 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-02, Section: A.
주기사항  
Advisor: Wang, Yue.
학위논문주기  
Thesis (Ph.D.)--The University of North Carolina at Chapel Hill, 2025.
초록/해제  
요약Explainable AI (XAI) research has proposed a variety of approaches for explaining machine learning model predictions to end-users. Feature importance explanation, which highlights input features that are most influential to the output, is a popular and effective approach. Studies have developed different algorithms to identity important features, and also empirically shown that such explanations can improve end-users' decision-making performance and understanding of the AI system in certain tasks. Despite these promising advancements, a critical gap remains open: features deemed predictive by algorithms can appear unintuitive to humans. For instance, prior studies found that the word ``problems'' predicts positive sentiment in product reviews, and the word ``Chicago'' is strongly associated with deceptive hotel reviews. However, most XAI research on feature importance explanations does not further explain why a feature is predictive. Therefore, this dissertation explores an underexplored area in XAI research: explaining unintuitive feature importance explanations. Using deception detection as a case study, this dissertation investigated two primary research questions.The first research question focused on developing a computational approach to explaining predictive but unintuitive words in deception detection. While previous research has used contextual information to explain unintuitive words in sentiment analysis, those in deception detection often represent some underlying phenomena that are too complex to be captured by local context. To this end, I leveraged a large language model (LLM) to conjecture the phenomena associated with words predictive of a review being genuine or deceptive, and then conducted algorithmic evaluations to validate the technical reliability of this LLM-based approach. The evaluation results confirmed that the LLM used in this dissertation (GPT-4o) generated non-hallucinated and generalizable phenomena that effectively explained why a word is predictive of genuine or deceptive reviews.The second research question focused on evaluating how unintuitive words and LLM-generated explanations influence participants in a decision-making task. To address this research question, I conducted a crowdsourced user study (N=220). In the study, participants first interacted with an AI system and judged a sequence of eight Chicago hotel reviews. Participants were randomly assigned to one of five interface conditions (i.e., a between-subjects design). The interface conditions varied in terms of the AI assistance features made available to participants. Then, participants independently completed a task of detecting deceptive hotel reviews from other cities. The study results found that showing predictive words without explaining why they are predictive was no better than not showing them at all, while explaining why these words are predictive significantly enhanced participants' learning of the task, appropriate reliance on AI assistance, and perceptions of the AI system.In summary, this dissertation makes three key contributions: (1) it sheds light on the critical issue of unintuitive feature importance explanations and the urgent need for making them understandable to end-users; (2) it introduces a conjecture-then-validate pipeline for explaining unintuitive words through LLMs; (3) it provides empirical insights into how unintuitive words and LLM-generated explanations influence end-users during decision-making. This dissertation offers an important design implication for future XAI research: machine-generated explanations should be aligned with human intuition and common sense to facilitate effective human-AI interaction.
일반주제명  
Information science
키워드  
Explainable AI
키워드  
Human-AI interaction
키워드  
Large language models
키워드  
Unintuitive features
키워드  
User studies
기타저자  
The University of North Carolina at Chapel Hill Information and Library Science
기본자료저록  
Dissertations Abstracts International. 87-02A.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2025        us                              c    eng  d
■001000017357333
■00520260202103209
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798291558157
■035    ▼a(MiAaPQ)AAI32000696
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a020
■1001  ▼aQu,  Jiaming.
■24510▼aExplaining  Unintuitive  Feature  Importance  Explanations
■260    ▼a[Sl]▼bThe  University  of  North  Carolina  at  Chapel  Hill▼c2025
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2025
■300    ▼a166  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-02,  Section:  A.
■500    ▼aAdvisor:  Wang,  Yue.
■5021  ▼aThesis  (Ph.D.)--The  University  of  North  Carolina  at  Chapel  Hill,  2025.
■520    ▼aExplainable  AI  (XAI)  research  has  proposed  a  variety  of  approaches  for  explaining  machine  learning  model  predictions  to  end-users.  Feature  importance  explanation,  which  highlights  input  features  that  are  most  influential  to  the  output,  is  a  popular  and  effective  approach.  Studies  have  developed  different  algorithms  to  identity  important  features,  and  also  empirically  shown  that  such  explanations  can  improve  end-users'  decision-making  performance  and  understanding  of  the  AI  system  in  certain  tasks.  Despite  these  promising  advancements,  a  critical  gap  remains  open:  features  deemed  predictive  by  algorithms  can  appear  unintuitive  to  humans.  For  instance,  prior  studies  found  that  the  word  ``problems''  predicts  positive  sentiment  in  product  reviews,  and  the  word  ``Chicago''  is  strongly  associated  with  deceptive  hotel  reviews.  However,  most  XAI  research  on  feature  importance  explanations  does  not  further  explain  why  a  feature  is  predictive.  Therefore,  this  dissertation  explores  an  underexplored  area  in  XAI  research:  explaining  unintuitive  feature  importance  explanations.  Using  deception  detection  as  a  case  study,  this  dissertation  investigated  two  primary  research  questions.The  first  research  question  focused  on  developing  a  computational  approach  to  explaining  predictive  but  unintuitive  words  in  deception  detection.  While  previous  research  has  used  contextual  information  to  explain  unintuitive  words  in  sentiment  analysis,  those  in  deception  detection  often  represent  some  underlying  phenomena  that  are  too  complex  to  be  captured  by  local  context.  To  this  end,  I  leveraged  a  large  language  model  (LLM)  to  conjecture  the  phenomena  associated  with  words  predictive  of  a  review  being  genuine  or  deceptive,  and  then  conducted  algorithmic  evaluations  to  validate  the  technical  reliability  of  this  LLM-based  approach.  The  evaluation  results  confirmed  that  the  LLM  used  in  this  dissertation  (GPT-4o)  generated  non-hallucinated  and  generalizable  phenomena  that  effectively  explained  why  a  word  is  predictive  of  genuine  or  deceptive  reviews.The  second  research  question  focused  on  evaluating  how  unintuitive  words  and  LLM-generated  explanations  influence  participants  in  a  decision-making  task.  To  address  this  research  question,  I  conducted  a  crowdsourced  user  study  (N=220).  In  the  study,  participants  first  interacted  with  an  AI  system  and  judged  a  sequence  of  eight  Chicago  hotel  reviews.  Participants  were  randomly  assigned  to  one  of  five  interface  conditions  (i.e.,  a  between-subjects  design).  The  interface  conditions  varied  in  terms  of  the  AI  assistance  features  made  available  to  participants.  Then,  participants  independently  completed  a  task  of  detecting  deceptive  hotel  reviews  from  other  cities.  The  study  results  found  that  showing  predictive  words  without  explaining  why  they  are  predictive  was  no  better  than  not  showing  them  at  all,  while  explaining  why  these  words  are  predictive  significantly  enhanced  participants'  learning  of  the  task,  appropriate  reliance  on  AI  assistance,  and  perceptions  of  the  AI  system.In  summary,  this  dissertation  makes  three  key  contributions:  (1)  it  sheds  light  on  the  critical  issue  of  unintuitive  feature  importance  explanations  and  the  urgent  need  for  making  them  understandable  to  end-users;  (2)  it  introduces  a  conjecture-then-validate  pipeline  for  explaining  unintuitive  words  through  LLMs;  (3)  it  provides  empirical  insights  into  how  unintuitive  words  and  LLM-generated  explanations  influence  end-users  during  decision-making.  This  dissertation  offers  an  important  design  implication  for  future  XAI  research:  machine-generated  explanations  should  be  aligned  with  human  intuition  and  common  sense  to  facilitate  effective  human-AI  interaction.
■590    ▼aSchool  code:  0153.
■650  4▼aInformation  science
■653    ▼aExplainable  AI
■653    ▼aHuman-AI  interaction
■653    ▼aLarge  language  models
■653    ▼aUnintuitive  features
■653    ▼aUser  studies
■690    ▼a0723
■71020▼aThe  University  of  North  Carolina  at  Chapel  Hill▼bInformation  and  Library  Science.
■7730  ▼tDissertations  Abstracts  International▼g87-02A.
■790    ▼a0153
■791    ▼aPh.D.
■792    ▼a2025
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17357333▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

Preview

Export

ChatGPT Discussion

AI Recommended Related Books


    New Books MORE
    Statistics for the past 3 years. Go to brief

    ค้นหาข้อมูลรายละเอียด

    • จองห้องพัก
    • ไม่อยู่
    • โฟลเดอร์ของฉัน
    • ขอดูแรก
    • Non-Book Loan Application
    • Nighttime Book Loan Application
    วัสดุ
    Reg No. Call No. ตำแหน่งที่ตั้ง สถานะ ยืมข้อมูล
    TF17710 전자도서 대출가능 My Folder 부재도서신고 비도서대출신청 야간 도서대출신청

    * จองมีอยู่ในหนังสือยืม เพื่อให้การสำรองที่นั่งคลิกที่ปุ่มจองห้องพัก

    Books borrowed together with this book

    Related Popular Books

    Available after logging in.