图书情报工作

图书情报工作 >

2024 , Vol. 68 >Issue 14: 14 - 24

DOI: https://doi.org/10.13266/j.issn.0252-3116.2024.14.002

理论研究

AIGC产品用户需求特征体系构建及改进策略研究

  • 毕达天 ,
  • 王璐 ,
  • 王雨菲 ,
  • 车尧
展开
  • 1 吉林大学商学与管理学院 长春 130012;
    2 中国科学技术信息研究所 北京 100038
毕达天,教授,博士;王璐,硕士研究生;王雨菲,硕士研究生;车尧,副编审,博士,通信作者,E-mail:chey@istic.ac.cn。

收稿日期: 2023-12-28

  修回日期: 2024-03-28

  网络出版日期: 2024-07-30

基金资助

本文系国家社会科学基金项目“基于用户跨社交媒体的信息行为偏好特征挖掘与推荐”（项目编号:21BTQ059）研究成果之一。

收起

Research on the Construction of User Demand Characteristic System and Improvement Strategy of AIGC Products

  • Bi Datian ,
  • Wang Lu ,
  • Wang Yufei ,
  • Che Yao
Expand
  • 1 School of Business and Management, Jilin University, Changchun 130012;
    2 Institute of Scientific and Technical Information of China, Beijing 100038

Received date: 2023-12-28

  Revised date: 2024-03-28

  Online published: 2024-07-30

Supported by

This work is supported by the National Social Science Fund of China project titled “Mining and recommendation based on users’ information behavior preferences across social media”(Grant No.21BTQ059).

Fold

摘要

[目的/意义] 从用户视角出发，探索AIGC产品的用户需求及特征，为国产AIGC产品质量的提升提供参考和建议，从而推动我国AIGC行业的高水平发展。[方法/过程] 以ChatGPT和文心一言为研究对象，爬取微博、知乎、抖音、百度贴吧4个社交媒体平台的116 046条评论文本，通过BERTopic主题模型提取用户关注主题，构建AIGC产品的用户需求特征体系，基于IPA分析工具从主题关注度和满意度两个维度分析国内外AIGC产品的需求特征差异，据此提出我国AIGC产品的改进策略。[结果/结论] AIGC产品的用户需求主要分为可获得性、可靠性、体验性、创新性、功能性、理解性、趣味性、安全性8类；我国AIGC产品的改进策略包括提升内容可靠性、增强文化适配度、完善安全保障体系、助推核心技术创新。

关键词： AIGC; 需求特征体系; 主题挖掘; BERTopic模型; IPA分析

本文引用格式

毕达天 , 王璐 , 王雨菲 , 车尧 . AIGC产品用户需求特征体系构建及改进策略研究[J]. 图书情报工作, 2024 , 68(14) : 14 -24 . DOI: 10.13266/j.issn.0252-3116.2024.14.002

Abstract

[Purpose/Significance] This paper aims to explore the user demand and characteristics of AIGC products from the perspective of user, providing reference and suggestions for the improvement of AIGC products quality, and promoting the high-level development of AIGC industry. [Method/Process] Taking ChatGPT and Wenxin as the research objects, this paper crawled 116 046 comment text from four social media platforms, including Weibo, Zhihu, Douyin and Baidu Tieba, then used BERTopic to extract the users’ focus topics and constructed a user demand characteristic system of AIGC products. Based on the IPA analysis tool, it analyzed the differences in demand characteristics of AIGC products at home and abroad from both attention and satisfaction dimensions, and it proposed the improvement strategy of AIGC products in China. [Result/Conclusion] The user demand of AIGC products are mainly divided into eight categories: availability, reliability, experience, innovation, functionality, understanding, interest and security. The improvement strategy for AIGC products in China includes improving the reliability of content, enhancing cultural adaptability, improving the security system, and boosting core technology innovation.

Key words： AIGC; demand characteristic system; topic mining; BERTopic model; IPA analysis

参考文献

[1] 量子位智库.中国AIGC产业全景报告[EB/OL].[2024-05-11]. https://new.qq.com/rain/a/20230410A01QZP00. (QbitAI Insights. Panoramic report of AIGC industry in China[EB/OL].[2024-05-11]. https://new.qq.com/rain/a/20230410A01QZP00.)
[2] 俞鼎,李正风.生成式人工智能社会实验的伦理问题及治理[J].科学学研究, 2024, 42(1):3-9.(YU D, LI Z F. Ethical issues and governance in generative artificial intelligence social experiment[J]. Studies in science of science, 2024, 42(1):3-9.)
[3] DALALAH D, DALALAH O M A. The false positives and false negatives of generative AI detection tools in education and academic research:the case of ChatGPT[J]. The international journal of management education, 2023, 21(2):100822.
[4] 刘亚丽,范逢春. ChatGPT-AIGC用户风险感知维度识别与治理研究--基于扎根理论的探索性分析[J].情报理论与实践, 2024, 47(3):121-129.(LIU Y L, FAN Y C. ChatGPT-AIGC users risk perception dimension identification and management research:an exploratory analysis based on grounded theory[J]. Information studies:theory&application, 2024, 47(3):121-129.)
[5] 中国信息通信研究院,京东探索研究院.人工智能生成内容(AIGC)白皮书[R/OL].[2024-05-11]. http://www.caict.ac.cn/english/research/whitepapers/202211/P020221111501862950279.pdf. (China Academy of Information and Communications Technology, JD Explore Academy. Artificial Intelligence Generated Content (AIGC) White Paper[R/OL].[2024-05-11]. http://www.caict.ac.cn/english/research/whitepapers/202211/P020221111501862950279.pdf.)
[6] STAHL B C, EKE D. The ethics of ChatGPT-Exploring the ethical issues of an emerging technology[J]. International journal of information management, 2024, 74:102700.
[7] 莫祖英,盘大清,刘欢,等.信息质量视角下AIGC虚假信息问题及根源分析[J].图书情报知识, 2023, 40(4):32-40.(MO Z Y, PAN D Q, LIU H, et al. Analysis on AIGC false information problem and root cause from the perspective of information quality[J]. Documentation, information&knowledge, 2023, 40(4):32-40.)
[8] 刘江峰,刘雏菲,齐月,等. AIGC助力数字人文研究的实践探索:SikuGPT驱动的古诗词生成研究[J].情报理论与实践, 2023, 46(5):23-31.(LIU J F, LIU C F, QI Y, et al. A practical exploration of AIGC-Powered digital humanities research:a SikuGPT driven research of ancient poetry generation[J]. Information studies:theory&application, 2023, 46(5):23-31.)
[9] 储节旺,罗怡帆,李佳轩. AIGC对信息生成方式及用户信息行为的影响[J].图书情报工作, 2023, 67(24):13-23.(CHU J W, LUO Y F, LI J X. The influence of AIGC on information generation mode and user information behavior[J]. Library and information service, 2023, 67(24):13-23.)
[10] DUONG C D, VU T N, NGO T V N. Applying a modified technology acceptance model to explain higher education students'usage of ChatGPT:a serial multiple mediation model with knowledge sharing as a moderator[J]. The international journal of management education, 2023, 21(3):100883.
[11] 卢新元,张进澳,雷晓鹏.人工智能生成内容环境下用户信息行为研究--以对话式搜索引擎为例[J].情报理论与实践, 2023, 46(12):84-92.(LU X Y, ZHANG J A, LIE X P. Research on user's information behavior in artificial intelligence generated content environment:take a conversational search engine as an example[J]. Information studies:theory&application, 2023, 46(12):84-92.)
[12] 毛太田,汤淦,马家伟,等.人工智能生成内容(AIGC)用户采纳意愿影响因素识别研究--以ChatGPT为例[J/OL].情报科学:1-15[2024-05-11]. http://kns.cnki.net/kcms/detail/22.1264.G2.20231103.1005.010.html. (MAO T T, TANG G, MA J W, et al. Factors influencing user adoption intention of artificial intelligence generated content (AIGC):a study on ChatGPT[J/OL]. Information science:1-15[2024-05-11]. http://kns.cnki.net/kcms/detail/22.1264.G2.20231103.1005.010.html.)
[13] 张莉曼,张向先,吴雅威,等.基于语义主题图谱的学术APP用户信息需求发现研究[J].情报理论与实践, 2021, 44(12):133-140.(ZHANG L W, ZHANG X X, WU Y W, et al. Research on information requirement discovery of academic APP users based on semantic topic map[J]. Information studies:theory&application, 2021, 44(12):133-140.)
[14] 夏苏迪,邓胜利,汪璠.在线医疗社区健康科普知识供需匹配研究[J].现代情报, 2023, 43(7):38-47.(XIA S D, DENG S L, WANG F. Research on supply optimization of health science knowledge in online medical community based on user need[J]. Journal of modern information, 2023, 43(7):38-47.)
[15] EUNHYE P, JINAH P, MINGMING H. Tourism demand forecasting with online news data mining[J]. Annals of tourism research, 2021, 90:103273.
[16] 单晓红,孔维嘉,王蕊.社交媒体数据驱动的老年人智能化需求研究[J].情报理论与实践, 2022, 45(8):23-30.(SHAN X H, KONG W J, WANG R. Research on the intelligent needs of the elderly driven by social media data[J]. Information studies:theory&application, 2022, 45(8):23-30.)
[17] JIE W, NARISA Z. What consumer complaints should hoteliers prioritize?Analysis of online reviews under different market segments[J]. Journal of hospitality marketing management, 2023, 32(1):1-28.
[18] 王文娜,许正良,李贺,等.共享住宿平台用户关注主题挖掘研究--典型城市差异分析的视角[J].图书情报工作, 2023, 67(9):121-131.(WANG W N, XU Z L, LI H, et al. Research on users'focus topic mining of shared accommodation platform:from the perspective of difference analysis of typical cities[J]. Library and information service, 2023, 67(9):121-131.)
[19] GROOTENDORST M. BERTopic:neural topic modeling with a class-based TF-IDF procedure[J]. ArXiv preprint arXiv:2203. 05794, 2022.
[20] CONTRERAS K, VERBEL G, SANCHEZ J, et al. Using topic modelling for analyzing panamanian parliamentary proceedings with neural and statistical methods[C]//2022 IEEE 40th Central America and Panama Convention (CONCAPAN). Panama:IEEE, 2022:1-6.
[21] 刘洋,柳卓心,金昊,等.基于BERTopic模型的用户层次化需求及动机分析--以抖音平台为例[J].情报杂志, 2023, 42(12):159-167.(LIU Y, LIU Z X, JIN H, et al. User hierarchical need and motivation analysis based on BERTopic model:taking Douyin platform as an example[J]. Journal of intelligence, 2023, 42(12):159-167.)
[22] 王浩伟,汪璠,王秉琰.主题视角下生成式人工智能生成内容与用户生成内容的比较[J].情报理论与实践, 2023, 46(10):200-207, 199.(WANG H W, WANG F, WANG B Y. A comparison of generative Al-generated content and usergenerated content in the thematic perspective[J]. Information studies:theory&application, 2023, 46(10):200-207, 199.)
[23] WANG Z, CHEN J, CHEN J, et al. Identifying interdisciplinary topics and their evolution based on BERTopic[J]. Scientometrics, 2023:1-26.
[24] 高春玲,姜莉媛,董天宇.基于BERTopic模型的老年人健康信息需求主题演化研究--以新浪微博平台为例[J/OL].情报科学:1-16[2024-03-25]. http://kns.cnki.net/kcms/detail/22.1264.G2.20240128.1743.004.html. (GAO C L, JIANG L Y, DONG T Y. Thematic evolution of health information needs in the elderly based on BERTopic model[J/OL]. Information science:1-16[2024-03-21]. http://kns.cnki.net/kcms/detail/22.1264.G2.20240128.1743.004.html.)
[25] MARTILLA J A, JAMES J C. Importance-performance analysis[J]. Journal of marketing, 1977, 41(1):77-79.
[26] 孙玲玲,胡彦蓉,刘洪久.基于LSTM-LDA算法和IPA分析的在线品牌社群用户关注热点研究[J].情报杂志, 2021, 40(9):178-186.(SUN L L, HU Y R, LIU H J. LSTM and LDA fusion algorithm analysis of online brand community for users'focus hot spots[J]. Journal of intelligence, 2021, 40(9):178-186.)
[27] PRATT S, SUNTIKUL W, AGYEIWAAH E. Determining the attributes of gastronomic tourism experience:applying impact‐ range performance and asymmetry analyses[J]. International journal of tourism research, 2020, 22(5):564-581.
[28] 吴江,李秋贝,胡忠义,等.基于IPA模型的乡村旅游景区游客满意度分析[J].数据分析与知识发现, 2023, 7(7):89-99.(WU J, LI Q B, HU Z Y, et al. Analyzing tourist satisfaction of rural scenic attractions based on IPA model[J]. Data analysis and knowledge discovery, 2023, 7(7):89-99.)
[29] 蒋翠清,吕孝忠,段锐.基于主题模型的产品在线论坛主题演化分析[J].系统工程学报, 2019, 34(5):598-609.(JIANG C Q, LV X Z, DUAN R. Analyzing topic evolution of online product forum based on topic model[J]. Journal of systems engineering, 2019, 34(5):598-609.)
[30] 熊曙初,李轩,吴佳妮,等.基于有监督对比学习的文本情感语义优化方法研究[J/OL].数据分析与知识发现:1-18[2024-05-11]. http://kns.cnki.net/kcms/detail/10.1478.G2.20231116.0936.002.html. (XIONG S C, LI X, WU J N, et al. Research of text affective semantic optimization method based on supervised contrastive learning[J/OL]. Data analysis and knowledge discovery:1-18[2024-05-11]. http://kns.cnki.net/kcms/detail/10.1478.G2.20231116.0936.002.html.)
[31] JAIN P K, PAMULA R, SRIVASTAVA G. A systematic literature review on machine learning applications for consumer sentiment analysis using online reviews[J]. Computer science review, 2021, 41:100413.
[32] 马晓悦,孙铭菲.融合热点事件主题演化的民族文化扩散研究[J].图书情报工作, 2022, 66(3):106-117.(MA X Y, SUN M F. Research on the diffusion of natural culture integrating the theme evolution of hot topics[J]. Library and information service, 2022, 66(3):106-117.)
Options
文章导航

/

〈 〉