User Experience Measurement and Improvement Strategies for Library and Information Services in the Era of Smart Libraries

Yanbian University, Yanji City, 133000

Abstract

As artificial intelligence technology applications deepen, smart library services are transitioning from traditional models to intelligent transformation, with greater emphasis on personalized services to deliver high-quality user experiences. Currently, smart libraries have established new library and information service frameworks after digital transformation. AI technology applications, intelligent information retrieval, and resource recommendation mechanisms ensure service quality and improve efficiency. Faced with increasingly diverse user demands, scientific measurement and continuous optimization of user experience are essential to guarantee service relevance. This paper investigates user experience measurement and improvement strategies for library and information services in the era of smart libraries.

Keywords: Smart library era; Library and information services; Human-computer interaction design; User experience measurement; Improvement

Smart libraries have undergone transformation through artificial intelligence technology, big data technology, and other innovations, enabling automated borrowing functions and intelligent navigation. However, challenges remain, including low user engagement and homogenized services. To effectively address these issues and meet escalating user demands, libraries must provide convenient services that satisfy personalized needs while enabling sufficient interaction. Traditional approaches struggle to reflect experience deficiencies \cite{1}. Establishing a user experience measurement model can accurately identify service shortcomings in libraries, propose improvement strategies, and serve as a basis for optimizing library resources and innovating services, thereby promoting the advancement of smart libraries toward "user-centered" ecological services.

1.1 Personalized Recommendation

If libraries continue to apply traditional management models, users must physically visit the library to obtain relevant information when needed. Within the library, they must utilize classification-based information retrieval, and in some cases, require staff recommendations. This operational approach presents numerous inconveniences, fails to deliver personalized services, and forces users to spend considerable time on searches, which does not align with today's fast-paced lifestyle. To create an intelligent environment, libraries must fully leverage smart technologies and their advantageous functions to provide personalized recommendations based on actual user needs, thereby delivering quality experiences \cite{3}. Particularly through the application of advanced data analysis technologies, systems can deeply analyze users' reading records during operation, understand their interests, and make recommendations according to their preferences.

For instance, for a user who loves historical books and primarily searches for historical topics in the library system, the library will prioritize recommending related books. Simultaneously, the system automatically collects information, recording user operations and behaviors when they log into electronic resource platforms for browsing, thereby identifying their specific interests. As this type of data accumulates, the system employs complex algorithms during operation to obtain precise results and identify the books users need from vast collections of resources.

For example, smart libraries operate an RFID intelligent document retrieval subsystem that ensures seamless integration with the OPAC system, enabling intelligent positioning and navigation functions to plan optimal paths. The RFID intelligent document retrieval subsystem primarily adopts a Browse/Server architecture, allowing users to query information more efficiently. Readers can employ fuzzy search methods by entering approximate information in dialog boxes for title, author, subject terms, etc., enabling multi-level queries to obtain detailed book information with graphical displays that accurately locate books on shelves \cite{4}. Through the Web publishing system, a sub-link on the library website provides readers with suitable book retrieval methods, making library services more user-friendly. The system also functions as an IE plugin, achieving seamless integration with the library OPAC system (Figure 1 [FIGURE:1]: 3D Book Positioning).

Analyzing Figure 1, the book location can be determined based on query results, and relevant information can be obtained by selecting the view positioning option. All book positioning information is based on actual conditions, with library structure floor plans quickly generated in the background, allowing automatic modification of library structures for convenient librarian maintenance. Using query results as a basis and combining them with the actual floor plan, clicking view positioning pops up a 3D floor plan of the library, with red dots marking current book locations \cite{5}.

1.2 Intelligent Search

When users search for information in libraries, facing massive amounts of data, finding what they need is extremely difficult. Intelligent search functionality makes information acquisition easier, more efficient, and precise. Through intelligent search, users can search using natural language, allowing the system to automatically understand user intent and extract valuable information needed from massive databases. Users can select any information field—such as title, author, subject terms, ISBN, or publisher—through a dropdown menu in the search box on the library website homepage. For example, by entering the book title "Dream of the Red Chamber" and clicking "Search" on the right, relevant results for this book will appear. To obtain bibliographic information, clicking on the book title reveals details and collection information, such as call number, location, and status \cite{6}. Users can note this information to locate books in the stacks. During system operation, search results are also adjusted based on users' search history and behavioral patterns. If a user frequently searches for certain types of information, the system prioritizes displaying relevant information, providing a personalized experience.

1.3 Artificial Intelligence Analysis

As various tasks in smart libraries become intelligent, the application of artificial intelligence technology requires in-depth interpretation of user needs, and leveraging the advantageous functions of intelligent tools can quickly address these issues. AI technology applications can automatically collect data on user operational behaviors, constructing comprehensive and precise user profiles based on this information \cite{7}.

To obtain behavioral data, Python language is used in combination with web crawler frameworks and appropriate toolkits, requiring proper implementation methods. Using downloaders to acquire crawled data, the parsed information is stored in MySQL databases. The obtained user data is displayed on library websites and stored in databases, allowing users to access it anytime and even review their borrowing history, further verifying user behavior.

During the process of processing and analyzing user data, Python is applied to collect data and clean it according to actual needs, removing missing data and eliminating meaningless information, including redundant data, defective data, and other non-valuable information. If data formats have issues, they must also be eliminated to ensure data quality and usability.

During data analysis, various advanced technologies should be fully utilized. Data mining techniques extract valuable data, while machine learning techniques enable deep analysis of processed data. By analyzing various user-related data—such as basic user information, library visit patterns, reading behaviors, preferences, and technology usage—more accurate user profiles can be established.

Based on data analysis results, user profiles are established and displayed across multiple dimensions \cite{8}. By utilizing these user profiles, libraries can grasp user needs, understand their reading habits and preferences, determine whether this information correlates with age or profession, and compare differences in book type preferences among different users.

The key procedures are as follows: constructing user profiles, counting each user's borrowing frequency, average borrowing duration, and extracting user search keywords.

# Count borrowing frequency for each user borrow_count = data.groupby('user_id') ['book_id'].count().reset_index(name='borrow_count')
# Calculate average borrowing duration for each user average_borrow_duration = data.groupby('user_id') ['borrow_duration'].mean().reset_index(name='average_borrow_duration'
# Extract user search keywords search_keywords = data.groupby('user_id') ['search_keywords'].apply(lambda x: ' '.join(x)).reset_index()
# Merge user profile information user_profile pd.merge(borrow_count, average_borrow_duration, on='user_id')
user_profile pd.merge(user_profile, search_keywords, on='user_id')

Analyze user behavior, classify users based on borrowing frequency and average borrowing duration, analyze user search keywords, and identify popular keywords.

# Classify users based on borrowing frequency and average duration user_profile['borrow_frequency']= np.where(user_profile['borrow_count']> user_profile['borrow_count'].median()，‘高频率"，‘低频率")
user_profile['borrow_duration_category']= np.where(user_profile['average_borrow_duration' ] > user_profile['average_borrow_duration' ].median()，"长时长"，“短
# Analyze user search keywords to identify popular keywords from collections import Counter
all_keywords =''.join(user_profile['search_keywords']).split()
keyword_counts = Counter(all_keywords)
top_keywords = keyword_counts.most_common(10)

2. Improvement Strategies

2.1 Enhancing Human-Computer Interaction Design

Human-computer interaction design enables all library equipment and users to interact on a specific platform where users receive friendly assistance, making borrowing more convenient and delivering quality experiences. A good human-computer interface should have clear content, simple operations, and intuitively presented functions \cite{10}.

Launching intelligent navigation and self-service systems, indoor positioning technologies such as Bluetooth beacons are applied to construct three-dimensional navigation systems, allowing readers to locate books in real-time through mobile apps without being physically present at the library, thereby reducing book-finding time. By deploying self-service borrowing and returning robots and 24-hour intelligent book cabinets that support facial recognition/QR code borrowing, waiting times for manual services are shortened.

In multimodal interaction experiences, voice assistants are introduced. For example, DeepSeek technology and similar solutions answer consultation questions and support natural language queries of collection resources. Additionally, combined with VR/AR technology, immersive reading scenarios can be provided, such as ancient book restoration demonstrations and historical scene reconstructions.

2.2 Personalized Information Services

By fully utilizing modern information technology, customized reading services are provided for users, recommending needed books or materials, delivering more accurate and reliable search results, and displaying required content. Big data analytics helps libraries locate materials of interest to users, clarify their needs, and grasp user behavior patterns, thereby providing personalized information services with stronger relevance and better quality \cite{12}.

Regarding user profiles and intelligent recommendations, based on data such as borrowing records and search keywords, machine learning algorithms generate user interest tags to proactively push relevant books and academic trends. Contextualized intelligent search optimizes search engine semantic understanding capabilities, supporting fuzzy queries (such as "rural revitalization case collections") and associatively recommending derived resources like policy documents and local practices. Cross-database resource integration is implemented, enabling one-click access to full-media content including e-books, audio-visual materials, and journal articles.

2.3 Smart Library Culture Construction

During the cultural construction of smart libraries, the focus extends beyond technology application to management model innovation. Libraries use information technology to shape an open environment and foster a shared cultural atmosphere. Various activities can accomplish this goal, such as organizing cultural events to disseminate smart library culture while users acquire relevant knowledge to better accept services \cite{14}. For example, libraries can collaborate with universities to host technology lectures by experts, inviting scholars to guide library work and introduce the latest information technology, enhancing staff knowledge for practical application. Libraries can also hold innovation competitions to increase user attention to library services, encouraging full interaction between staff and users, focusing on user ideas for service innovation, and developing innovative service plans based on these ideas. These activities enliven the library's cultural atmosphere, stimulate librarians' innovation enthusiasm, foster user awareness of service innovation, and encourage active participation in smart library construction.

For cultural inheritance and community connection, digital technologies are utilized to revitalize rural intangible cultural heritage resources, such as establishing dialect voice databases and folk culture digital exhibition halls to enhance cultural identity. To create a library learning ecosystem, "Reading Growth Profiles" are created to record user learning trajectories and generate competency analysis reports (such as foreign language reading advancement suggestions). Thematic reading challenge competitions can also be established to enhance user engagement through point-based reward mechanisms \cite{15}.

2.4 Smart Governance System Construction

Regarding organizational structure, libraries should adjust from a command perspective, requiring analysis of current departmental setups and personnel division. The responsibilities, work content, and methods of each department must be clarified, along with understanding all staff members' authority. For example, libraries can establish a dedicated information technology department for information management, staffed with personnel who maintain library intelligent systems, possess data management capabilities, and can drive technological innovation based on actual library operations. Additionally, a user service department is needed to provide required services, handle user consultations, organize feedback, and resolve user complaints.

For example, libraries applying the Qingdao Hengrui patent system can monitor real-time foot traffic and book circulation status across branches, automatically recommending the nearest available branch and alternative titles during main library maintenance. Through cloud computing platforms, regional inter-library resource allocation is achieved, avoiding long-term shortages of popular books. Data-driven decision optimization analyzes user behavior big data (such as shelf occupancy rates and high-frequency search terms) to guide procurement strategies and service adjustments. A service quality KPI system is established (including metrics like borrowing efficiency improvement rate and user satisfaction) and integrated into a smart management dashboard.

Through this research, it is clear that to achieve sustainable development, smart libraries must attach great importance to user experience, which serves as both the starting point and the ultimate goal of library service intelligent transformation. Smart libraries can better meet user needs while improving user experience of library and information services, driving library services toward greater intelligence, personalization, and efficiency. In the future, as technology continues to advance and user needs evolve, smart libraries will continue to innovate and develop, bringing greater convenience to users. Libraries will no longer be merely places for book storage but will become vibrant, innovative knowledge service centers and cultural exchange platforms. This study proposes that smart libraries apply artificial intelligence technology to provide users with quality experience environments, utilize generative AI to enhance interaction naturalness, and promote the improvement of intelligent knowledge service platforms. Smart libraries are not only knowledge storage spaces but also core hubs for stimulating creativity and promoting lifelong learning. Only by continuously focusing on user experience, providing personalized services, and balancing technology with humanism can smart libraries maximize their social value.

References

[1] Duan Meizhen, Chu Jingli, Zhang Dongrong, et al. Investigation and Analysis on the Construction Status of Smart Libraries in "Double First-Class" Universities[J]. Library Tribune, 2022, 042(1):91-101.

[2] Li Lirui, Zhang Jiacheng, Deng Xiaozhao. Research on the Influence Mechanism of User Experience of Library Robot Services[J]. Library and Information Service, 2024, 068(12):43-55.

[3] Ma Kunkun, Zhu Lingling, Mao Yihong. Analysis of the Current Situation of Smart Reading Services on Academic Platforms and the Empowerment Path of AIGC Technology[J]. Library and Information Service, 2025, 069(2):45-55.

[4] Cheng Xiufeng, Cao Qi, Jiang Kaichen, et al. Research on Technical Element Analysis and Knowledge Graph Construction in the Process of Smart Library Construction[J]. Library and Information Service, 2024, 068(9):123-136.

[5] Zheng Dejun, Huang Peng, Li Yang. Evaluation Model of User Participation Level in Subject Services of University Libraries—From the Perspective of High-Quality User Participation[J]. Library and Information Service, 2022, 066(12):9-10.

[6] Anonymous. Libraries and New Quality Productivity: Scenario-Driven Data Governance and Artificial Intelligence—Notice for the 4th Smart Library and Smart Service Academic Symposium (Third Round)[J]. Library and Information Service, 2024, 068(18):147-148.

[7] Hu Juan, Ke Ping. Research on the Smart Choices and Development Strategies of Library Main-Branch Systems in the Construction of Smart Library Systems[J]. Library and Information Service, 2023, 067(5):9-10.

[8] Chen Guanting, Zhang Zhen, Huang Qi. Smart Libraries from the Perspective of the Metaverse: A Service Ecology Integrating Human Wisdom and Object Intelligence[J]. Library and Information Service, 2023, 067(10):15-25.

[9] Ran Haihe. Research on Smart Library Construction from the Perspective of Smart Cities—A Review of "Smart City and Smart Library"[J]. Modern Urban Research, 2023, 038(3):5-10.

[10] Du Xilin, Wang Yu, Che Baojing. Theoretical Reflections and Implementation Paths for Innovation in Public Library Service Systems Empowered by Digital Intelligence[J]. Library and Information Service, 2024, 068(7):3-10.

[11] Wen Mingzhang, Qian Guofu. Comparative Study on User Behavior Data of Electronic Resources in University Libraries Based on Different Client Software Collection Technologies[J]. Library and Information Service, 2022, 66(6):11-12.

[12] Xiang Shang, Sun Zhao, Yuan Lingjian, et al. Research on the Identification and Development Trends of Related Technologies for Smart Libraries in China from the Perspective of Patent Data Mining[J]. Library and Information Service, 2024, 068(19):78-89.

[13] Shen Xiang, Yu Gaihong, Wang Li, et al. Research and Practice of Scientific and Technological Strategic Intelligence Based on DIKW Theory—A Case Study of the Information Portal Service for Key Integrated Circuit Fields at the Documentation and Information Center of the Chinese Academy of Sciences[J]. Library and Information Service, 2025, 069(9):29-36.

[14] Wang Xinyu, Wang Yi, Shao Bo. Reconstruction Scheme and Smart Implementation Path of Library Consortia Based on Next-Generation Service Platforms[J]. Library and Information Service, 2025, 069(9):90-99.

[15] Yang Jing, He Cong, Wei Jixun. Construction of Library Knowledge Service System Under the Background of First-Class University Construction—A Case Study of Beijing Institute of Technology Library[J]. Library and Information Service, 2023, 067(23):38-48.

[16] Ding Qiujing, Su Jing. Research on Optimization Strategies for Library Information Resource Construction Oriented Toward Generative Artificial Intelligence (AIGC)[J]. Library and Information Service, 2024, 068(18):23-31.