Factors Influencing User Satisfaction with Artificial Intelligence Applications: An Integrated Approach of Semantic Network and BP Neural Network
-- A Case Study of DouBao APP
DOI:
https://doi.org/10.54097/z6caqg52Keywords:
IWE-AS Model, Semantic Network, LDA Model, BP Neural Network, Online Reviews, User SatisfactionAbstract
Generative AI has moved quickly from laboratory demos to everyday mobile apps in China, and user satisfaction has become the clearest test of whether these products actually deliver value. Yet most satisfaction studies still lean on questionnaires — useful, but too structured to capture what people really complain about or praise after prolonged use. This paper looks instead at what users themselves wrote. We collected 63,032 online reviews of ByteDance's DouBao APP (54,248 positive and 8,784 negative, November 2024 to October 2025, compiled via DianDian Data) and asked a simple question: what drives satisfaction, and how much does each driver matter? Rather than relying on a single method, we combined four steps. First, an IWE-AS keyword extraction model — which adds position encoding and a TDF adjustment to the classic TF-IDF — was used to surface terms that genuinely reflect concern, not just frequency. Semantic network mapping (ROSTCM6) and LDA topic modeling (with pyLDAvis for tuning) then revealed how those keywords cluster into themes. Finally, a BP neural network translated those themes into weights in a purely data-driven way, avoiding the subjectivity of expert scoring. The resulting evaluation system covers 6 primary dimensions and 24 secondary items. Interaction (0.347) and accuracy (0.238) dominate — together close to 60% of total weight — followed by intelligence (0.122), practicality (0.116), learning capability (0.101) and overall functionality (0.076). Within those, 'chat / picture generation' and 'failure to understand instructions' in the interaction dimension, and 'speechless / incorrect answers' in the accuracy dimension, carry the highest secondary weights. The implication is practical: for DouBao and similar domestic AI apps, improving how well the system understands colloquial instructions and how reliably it answers — especially for study-related queries — will move satisfaction far more than adding new peripheral features.
Downloads
References
[1] Li, X., & Wang, Y. (2024). The development trend of global large scale artificial intelligence models and their industrial impact. Journal of Artificial Intelligence Research, 62(1), 45–78. https://doi.org/10.1613/jair.1.15243.
[2] Ministry of Science and Technology of the People's Republic of China. (2023). Strategic guidelines for accelerating the development of a new generation of artificial intelligence. Science & Technology Review, 41(8), 5–18. https://doi.org/10.3969/j.issn.1000 7857.2023.08.001.
[3] Oliver, R. L. (1980). A cognitive model of the antecedents and consequences of satisfaction decisions. Journal of Marketing Research, 17(4), 460–469. https://doi.org/10.1177/002224378001700405.
[4] Geng, X. L., & Zhang, H. (2024). Exploring user satisfaction factors of e commerce products based on text mining: A case study of JD Apple mobile phones. Computational Economics, 63(3), 1123–1145. https://doi.org/10.1007/s10614 023 10324 x.
[5] Yan, X. Y., Liu, J., & Chen, Z. (2023). Quantifying user satisfaction of charging stations via K Means clustering and BP neural network. Sustainable Cities and Society, 98, 104789. https://doi.org/10.1016/j.scs.2023.104789.
[6] Chen, Z. J., & Wang, Q. (2024). Analyzing the contribution of product attributes to user satisfaction: Evidence from JD mobile phone reviews. Electronic Commerce Research and Applications, 58, 101782. https://doi.org/10.1016/j.elerap.2024.101782.
[7] Feng, Y., & Zhao, M. (2023). Constructing a multi dimensional satisfaction evaluation system for online health communities using BERTopic and LSTM. Journal of Medical Internet Research, 25(6), e43215. https://doi.org/10.2196/43215.
[8] Guo, T. Y., & Li, S. (2024). Optimizing TF IDF for text classification: Integrating chi square statistics and position weight. Pattern Recognition Letters, 178, 109–117. https://doi.org/10.1016/j.patrec.2024.02.005.
[9] Sun, G. Y., & Yang, H. (2023). IWE AS: A text classification model based on improved word embedding and adaptive segmentation. IEEE Transactions on Knowledge and Data Engineering, 35(12), 11890–11902. https://doi.org/10.1109/TKDE.2023.3289456.
[10] Maisimujiang, N., & Abudureyimu, A. (2024). A keyword extraction method integrating BERT, LDA, and TextRank for user review analysis. Neural Computing and Applications, 36(8), 6215–6232. https://doi.org/10.1007/s00521 023 08876 9.
[11] Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent Dirichlet allocation. Journal of Machine Learning Research, 3, 993 1022. https://doi.org/10.5555/944919.944940.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Frontiers in Computing and Intelligent Systems
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
