Ebook: Disruptive Human Resource Management

The concept of Disruptive Human Resource Management (DHRM) has emerged in today’s fast-paced digital era, blending people, technology, and strategy to reshape organizational landscapes. Human Resource Management (HRM) has undergone a significant transformation over the years, shifting from what was traditionally an administrative function to become a strategic partner in business growth. At the heart of disruptive HRM lies an emphasis on people. Employees are no longer viewed merely as a resource but as a valuable asset capable of driving innovation and growth. This shift towards a people-centric approach recognizes the importance of employee engagement, development, and wellbeing in achieving organizational success.

Technology plays a major role in disruptive HRM, enabling organizations to streamline processes, enhance communication, and make data-driven decisions. Advanced HR technologies such as Artificial Intelligence (AI), Machine Learning (ML), and data analytics offer insights into employee performance, preferences, and behaviors, empowering HR professionals to tailor strategies that meet individual and organizational needs.

Furthermore, technology can facilitate remote work, flexible scheduling, and digital learning opportunities, inculcating a more inclusive and adaptable work environment. This digital transformation not only enhances operational efficiency but also promotes collaboration, creativity, and agility within teams. Strategic alignment is crucial in disruptive HRM, ensuring that HR initiatives are closely integrated with organizational goals and objectives. By aligning people practices with business strategies, HR can drive innovation, improve competitiveness, and create sustainable value for all.

Disruptive HRM encourages a culture of continuous learning and adaptability, preparing employees to accept change and deal with uncertainties. It emphasizes the importance of upskilling and reskilling, empowering employees to acquire new competencies and stay relevant in evolving industries. Moreover, disruptive HRM prioritizes diversity, equality, and inclusion (DEI), recognizing the unique contributions of a diverse workforce and fostering a culture of belonging. By embracing DEI initiatives, organizations can harness the power of diverse perspectives, drive innovation, and build stronger connections with customers and communities. The purpose the International Conference on Disruptive Human Resource Management: People, Technology, and Strategy is to bring together leading thinkers, HR professionals, academics, and experts from industry to explore innovative approaches, share cutting-edge research, and discuss best practice in leveraging people, technology, and strategic alignment to drive organizational success in an ever-changing global landscape. Attendees from across the globe brought with them a diverse array of perspectives, experiences, and insights, thereby enriching the discourse and expanding the knowledge horizons of this rapidly evolving field.

The conference represents a paradigm shift in how organizations manage their most valuable asset – their people. By integrating people, technology, and strategy, disruptive HRM enables organizations to adapt to a changing market dynamic, drive organizational adaptability, and introduce a culture of innovation and excellence. Embracing disruptive HRM is essential for those organizations aiming to survive in today’s complex and competitive business landscape.

On behalf of the DHRM 2024 organizing committee, I am delighted and honored to welcome you to the International Conference on Disruptive Human Resource Management: People, Technology and Strategy. I believe in the purpose of Disruptive Human Resource Management: to transform traditional HR practices by integrating people, technology, and strategy to drive innovation, promote flexibility, and create sustainable value for organizations in a rapidly evolving business landscape.

Message from the Program Chair

Disruptive Human Resource Management is centered around the idea of utilizing disruptive technologies and innovative strategies to transform HR functions and create value for both employees and the organization. This approach goes beyond traditional HR practices focused solely on administrative tasks, compliance, and employee relations. Instead, DHRM emphasizes strategic alignment, flexibility, and continuous improvement, enabling organizations to anticipate and respond proactively to market shifts and industry disruptions. The key principles of DHRM are the integration of technology into HR processes, often referred to as HR Tech. Advanced technologies such as Artificial Intelligence (AI), Machine Learning (ML), and data analytics are revolutionizing HR functions by automating routine tasks, enhancing decision-making, and providing valuable insights into employee performance, engagement, and retention. This technological integration not only improves operational efficiency, but also enables HR professionals to focus on strategic initiatives, talent development, and organizational growth.

Above all, the primary feature of DHRM is its emphasis on employee experience and engagement. In the era of DHRM, employees are viewed as a valuable asset capable of driving innovation, productivity, and business success. Organizations adopting DHRM prioritize employee wellbeing, professional development, and career growth, creating a culture that fosters creativity, collaboration, and continuous learning. By investing in employee experience, organizations can attract top talent, reduce turnover, and build a motivated and engaged workforce poised for success.

The DHRM 2024 conference attracted an impressive array of submissions, with over 86 papers submitted from a wide range of research disciplines. This increase in the number of contributions reflects the dynamic momentum of advance in the field of Disruptive Human Resource Management. After a rigorous and meticulous review process, 39 research papers were selected for their innovative approach, high quality, and originality. These have been organized into three distinct categories to showcase the diverse facets of DHRM: (i) Human Capital with Technical Space, (ii) Business Strategy, and (iii) People Management.

The first section, Human Capital with Technical Space, contains 18 contributions from esteemed researchers and showcases the category that forms the backbone of Disruptive Human Resource Management. The papers included here explore the intersection of human resource management and technology, emphasizing the critical role of employees in maximizing technological advancements for organizational success and examining how organizations can apply the potential of their workforce to adapt to technological change in order to innovate, and drive business growth in the digital era. Many valuable insights are shared by researchers, including positive guidance for college students in career planning values combined with computer data simulation analysis; age-appropriate web design for people over 60 with cognitive difficulties; research into innovative ways of talent-training mode combining computer-aided teaching with student management; analyzing the links between agricultural modernization and the livelihoods of rural residents: a case study from China; a simulation of an enterprise human resource-management model based on long short-term memory (LSTM); an optimization method for human-resource decisions; hybrid mechanism of deep q-network (DQN) and graph-mining algorithm in six-dimensional integrated mental-health assessment and prediction: taking Hefei University of Economics as an example; and the design and implementation of a GIS-based historical catering-culture evolution system and talent-competency-model prediction platform based on intelligent interactive personality-testing software and intelligent-cluster analysis

The next section, entitled Business Strategy, presents 10 selected papers in which subjects covered include the empirical impact of debt financing on the level of environmental information disclosure in heavily polluting enterprises; the application and development of AI in machinery manufacturing; blockchain technology and the improvement of ESG information transparency; a customer-sentiment and demand-prediction model based on cognitive computing; the optimization and prediction of risk factors in a centralized financial-management system based on a prediction algorithm; market segmentation and personalized marketing strategies; the construction of an internet-based cross-border E-commerce service platform; and the simulated design of a warehouse business process based on activity and WMS.

The concluding section of the proceedings features a compelling compilation of papers dedicated to People Management. It includes 11 research papers related to the application of big data analysis and prediction in hotel room pricing; analysis of a social public-management guarantee mode based on a big data algorithm; the design and implementation of an intelligent question-answering robot for human resources and social security based on DCU and ChatGLM6B models; case analysis and strategy research on service outsourcing in Changsha; the improvement and application of big-data analysis methods to traditional, international political forecasting; research into the influence of group-fitness activities on the quality of life of the elderly based on big-data analysis; the application of a career management and competency model in the construction of new employee training systems in enterprises; career path optimization of DQN in career planning for college students; group intelligence and psychodynamics: intelligent motivation and psychological factors in the group behavior of employees; scene reproduction and interactive experience based on virtual-reality technology in the study of ancient literature; and the prediction of and intervention in mental-health problems in college students: a temporal attention model based on a recurrent neural network.

The proceedings of DHRM 2024 stand as a compelling testament to the rapid evolution of the fields of disruptive technologies and innovative strategies to transform HR functions. By embracing disruptive technologies and strategies, DHRM empowers organizations to adapt swiftly to change and provide a culture of continuous improvement. This forward-thinking approach not only enhances organizational resilience, but also drives employee engagement and productivity. As businesses navigate an increasingly complex and dynamic landscape, DHRM is emerging as a crucial force in shaping the future of work. We look forward to re-assembling in the future to continue this collaborative journey towards utilizing the full potential of disruptive technologies and innovative strategies to transform HR and shape the future of work.

Acknowledgements

We are extremely grateful to the authors for their contributions to this volume, and to the reviewers who have contributed their time and expertise to maintain the quality of the conference and proceedings.

We would also like to thank IOS Press for their support in publishing the DHRM 2024 Conference Proceedings.

Last but not least we are extremely grateful to Interscience Research Network (IRNet) International for organizing the conference and providing the support and services required to organize this event successfully.

We are sure that the readers will derive immense benefit and knowledge from this volume, and we also look forward to your valued contributions and support for coming editions of the International Conference on Disruptive Human Resource Management: People, Technology and Strategy (DHRM).

Dr. Srikanta Patnaik, Director, IIMT, Bhubaneswar, India

Program Chair

Career is a key stage for a person to realize his own value. However, with the development of society and the continuous change of organizations, the traditional career stability is getting smaller and smaller, and career planning becomes more and more important. This paper combines computer data simulation technology to analyze college students’ career planning, and puts forward a structural optimization design method based on parallel computing and genetic algorithm. Moreover, this paper uses Python to realize the computer simulation analysis of college students’ career planning data, uses adaptive genetic algorithm as the data processing algorithm, and constructs the career planning system model from three dimensions: self-management, organizational support and environmental penetration. From the experimental evaluation results, it can be seen that the college students’ career planning system proposed in this paper has good results, can effectively guide college students’ study and life, and has an important role in promoting the cultivation of college students’ good values.

With the development of information technology and the aging of the world’s population, how to design websites that are easy to access and use for the elderly is becoming an increasingly important issue. However, due to the cognitive aging of the elderly in terms of vision, memory, and processing speed, they often face varying degrees of cognitive difficulties when browsing the web. The cognitive difficulty cause a significant negative impact on the user experience of the elderly browsing the web and hinders the integration of the elderly into the information era. In order to reduce the cognitive difficulties of the elderly when browsing the web, and to explore quantifiable principles of age-appropriate web design, This study manipulated four quantitative indicators: color, contrast, page layout and operation logic to design the web page, and then the elderly over 60 years old were invited to rate different web page designs through a questionnaire survey. Finally, a comprehensive mathematical model for aging web design was built based on the data from the questionnaire survey. The research results provide a quantitative idea and a mathematical model that integrates multiple indicators for aging-appropriate web design.

In order to promote the effective application of intelligent teaching in modern teaching, this paper combines big data technology to build an educational information platform, integrates intelligent education into actual teaching. Moreover, this paper uses educational data related technology to judge and analyze it through technical level, so as to enhance the effectiveness of education. Moreover, the evaluation method proposed in this paper has changed from traditional evaluation based on teachers’ impression and experience of a single individual to data-based evaluation. The way of teaching evaluation no longer only depends on teachers’ evaluation of individual impression, but also depends on education big data related technologies to find a law of teaching activities and better optimize and improve the teaching process. From the experimental results, we can see that the education information platform based on big data proposed in this paper has a good improvement on the traditional online teaching mode.

In order to promote the innovation and development of talent training mode combining computer-aided teaching with student management, this paper analyzes the talent training mode of colleges and universities with computer-aided technology, constructs an intelligent platform suitable for modern colleges and universities, analyzes the demand of college students’ talent training mode through demand analysis, constructs the system function module, fully analyzes the system operation principle and information transmission module, and puts forward the system construction scheme. Moreover, this paper analyzes the system from the aspects of personnel training, college examination and practical work, and obtains several structural modules of the system. In order to verify the practical effect of the system platform proposed in this paper, this paper tests the effect of the model proposed in this paper on talent training in colleges and universities through experimental teaching mode.

In the process of English translation, how to use monolingual corpus which is relatively easy to obtain and realize corpus expansion is the premise of applying intelligent term matching method. In order to improve the accuracy of automatic matching of technical terms in English translation, this paper proposes a semi-supervised learning translation method which incorporates the bottleneck of variational information. This method uses small-scale parallel corpus training to obtain the basic translation model and uses back translation to translate large-scale monolingual corpus into pseudo-parallel corpus, and then merges the two parallel corpus, so that the corpus scale can be well applied to variational method. The experimental results of semi-supervised learning on multiple translation data sets show that compared with the benchmark system, the proposed method can effectively improve the translation quality while maintaining the translation sentence length, and solve the problem of over-translation in traditional neural machine translation to a certain extent.

In order to improve the efficiency of English translation teaching, it is necessary to make changes in English teaching. This paper combines computer translation technology to change the traditional English teaching process, and selects the fundamental frequency as the acoustic parameter to reflect the mood information. The English teaching system combined with computer translation technology relies on the constantly updated video resource library, grasps the two key links of teaching and training, creates a situational teaching and training environment for students, and constructs a teaching mode based on intelligent multifunctional simultaneous interpretation laboratory. In order to verify the application effect of the intelligent multifunctional simultaneous interpretation laboratory, the system effect is verified by expert evaluation and student evaluation. Through experimental research, it can be seen that the English teaching method combined with computer translation technology proposed in this paper has good effect and can effectively promote the effective improvement of English translation teaching quality.

Agriculture is crucial for a country’s overall development, heavily impacting the national economy. In China’s modernization context, agricultural modernization plays a pivotal role by transforming traditional farming into modern agriculture through advances in mechanization, irrigation, and electrification. This study investigates the spatial and temporal patterns of rural residents’ income from 2005 to 2020. It explores the relationships between mechanization, irrigation, and electrification, and their impacts on residents’ livelihoods, including income and grain production from 2003 to 2023. Additionally, a ridge regression model is used to analyze developmental trends and factors that support rural residents’ livelihoods. Findings reveal significant variations in per capita income among rural residents across different provinces in China. The study highlights the importance of resource advantages, rural economic development, and government support. It also identifies a strengthening spatial clustering pattern in rural residents’ income over time. Moreover, a strong correlation is observed between income, grain production, and variables such as agricultural machinery power, effective irrigation area, and power generation capacity. Regression analysis confirms these relationships with an accuracy of over 0.95 (R²). This study employs statistical analysis and machine learning to explore and predict the impact of agricultural modernization on farmers’ livelihoods, providing insights into modern agricultural development. The results contribute to a better understanding of the complex dynamics between agricultural modernization, rural residents’ income, and related factors. Ultimately, this research aids in policy formulation and decision-making for national development.

Engineering management plays a crucial role in the successful execution of engineering projects by coordinating and controlling resources, planning, and ensuring timely delivery. However, with the increasing complexity of projects and the need for innovation and competitiveness, there is a growing demand for advanced methods to evaluate and analyze project performance. This paper proposes a novel approach using the particle swarm optimization backpropagation neural network for performance analysis in engineering management. In the first place, the improved particle swarm optimization (IPSO) algorithm that this study develops is an improvement on PSO with respect to both inertia weight and learning factor. Second, this paper employs IPSO technique to improve BP network to address the issue of poor convergence performance due to random initialization of network parameters. In conclusion, this work conducted a number of experiments on IPSO-BP, and the results of those experiments confirmed the method’s superiority. The integration of PSO and BP algorithms offers a promising solution to overcome the limitations of conventional methods. The findings of this research have practical implications for project managers, enabling them to make informed decisions and optimize project performance.

Human resource management is a comprehensive process involving recruiting, training, evaluating and motivating employees to achieve organizational goals. In today’s society, human resource management forecasts have low accuracy and high error rates. This article uses the BP algorithm to build a turnover and salary prediction model to predict human resource management. Finally, through experiments, it was concluded that the algorithm greatly reduced the error rate of human resource management predictions and improved its accuracy, with an accuracy rate as high as 98.9%. In the future, BP algorithm and human resource management prediction should be widely used.

As an enterprise HRM (Human Resources Management), how to realize the scientific and rational management of the enterprise with the help of BD (big data) technology, and how to mine the information that is beneficial to the decision-making of the enterprise from a large amount of data has become an important issue at present. This paper studies and designs HR (Human Resources) performance appraisal system with BD technology. The advanced nature, maintainability and expansibility of the system should be considered as much as possible in the design. This system adopts B/S structure and ASP.NET three-tier architecture mode. Aiming at a large number of uncertain management factors in HRM, this paper puts forward an HR information management model based on DT (Decision Tree) algorithm. Aiming at the shortcomings that the logical expression of ID3 algorithm is not clear enough and easy to understand, this paper puts forward a new idea of building a tree. In the process of tree building, when the proportion of objects in the subset belonging to the same class exceeds a certain probability, the growth of the tree is stopped, and the node is marked as a leaf node of most classes. After system testing, the correctness of the target system is verified.

With the rapid development of human resource management theory and the increasing improvement of management technology, enterprises choose solutions based on qualitative analysis by experienced experts when making decisions, and the solutions are relatively simple. The disadvantages of this traditional qualitative method are unscientific, imprecise, and costly. Therefore, this article has conducted research on the optimal decision-making of enterprise human resource planning through DM (Data Mining) algorithm. The experimental results show that using this algorithm to design a human resource planning decision support system improves the registration and timeliness of human resource scheduling and allocation, and has good human resource information management capabilities, which has certain application value in human resource information management. By constructing a human resource planning and decision support system, we can classify, store, and schedule human resource information, improve the information distribution and scientific coordination capabilities of human resources, thereby better exerting the initiative of human resources, promoting the development of post construction, and further improving production capacity.

Competency is an effective tool for human resource management research. Compared with the traditional recruitment, which lacks the prediction of talents’ future performance and creativity, applying competency model to recruitment is conducive to building the core competitiveness of modern enterprises. In contemporary times, to bolster their competitiveness, enterprises must prioritize effective human resource management. This entails refining the recruitment system and attracting top-tier talent using a competency-based model. This paper explores the utilization of the competency model in managing enterprise human resource recruitment. Furthermore, it introduces a talent evaluation approach that integrates fuzzy theory with NN (neural networks). This method uses fuzzy transformation to transform the fuzzy evaluation of hierarchical indicators into the overall fuzzy evaluation, and uses BPNN (Back propagation neural network) to calculate the final result corresponding to the overall fuzzy evaluation. The simulation results show that the MSE (Mean squared error) of this method is about 4.32, and the MAE (Mean absolute error) can reach 1.36, and the evaluation accuracy and reliability of the algorithm are high. It can improve the recruitment quality of human resources in enterprises and promote enterprises to have better development prospects.

This article aims to solve the problem of resource optimization in the management of scientific research projects in universities, and proposes a resource optimization model based on Genetic algorithm (GA). By collecting and processing real data and conducting simulation experiments, this article evaluates the performance and stability of the model. The experimental results show that GA has good convergence and solution quality in resource optimization. Compared with other methods, GA has advantages in optimization effect and stability. It can maintain good performance under different data sets and parameter settings, and its stability is basically around 89.5%. The stability test further verified the reliability of the model in different scenarios. Therefore, this article holds that GA has application potential in the optimization of scientific research project management resources in universities, which can provide effective decision support for managers and realize rational allocation and efficient utilization of resources. These results provide an important reference for future research, and can promote the development of scientific research project management in universities in a more intelligent and optimized direction.

The survival and development of enterprises depend on effective human resources, especially in the context of the increasing demand for simplified, scientific, and automated management in modern enterprises. This article explores the application of Long Short Term Memory (LSTM) algorithm in contemporary human resource (HR) management to address challenges such as low efficiency and mismatch between personnel and demand. By utilizing the sequence modeling ability of LSTM networks, the model can better capture the temporal characteristics and correlation relationships of HR data, thereby improving the accuracy and reliability of decision-making. The research results indicate that the LSTM algorithm has the smallest model error when the number of iterations is 11. Using LSTM algorithm in a certain time series can effectively predict the HR required by enterprises, achieve refined and automated HR management, thereby making correct decisions for the development of enterprises and improving work efficiency. The experimental results prove that LSTM can provide correct reference opinions for enterprise decision-making and has a good promoting effect on the development of enterprises.

When enterprises face a large amount of enterprise data, how to effectively utilize a large amount of information to better manage talent has become a key issue that urgently needs to be solved. Therefore, this article proposes a human resource decision-making optimization method based on the Hadoop big data platform. By taking steps such as data collection, feature extraction, model construction, and decision optimization, the scientificity and accuracy of enterprise human resource management can be improved. This article compares the performance of the algorithm used with machine learning (ML) and deep learning (DL) algorithms, and the results indicate that the F1 value of the method designed in this article is 95.8%, which is higher than 84.7% of DL and 77.9% of ML. Experiments have shown that this method can effectively enhance the accuracy and efficiency of human resource decision-making, providing strong support for human resource management in enterprises.

How to evaluate the mental health status of college students? How to establish evaluation indicators for the mental health status of college students? How to predict mental health status? These are important issues faced by mental health education in universities. In response to the above issues, the article proposed a hybrid mechanism based on Deep Q-Network (DQN) and graph mining algorithm, and applied it to the “Six-dimensional Integrated” mental health assessment and prediction at Hefei University of Economics. The experimental results show that the prediction method based on Deep Q-Network has the highest accuracy of 98%, the highest recall rate of 88%, and the highest stability of 96%. From this, it can be seen that the proposed mixed mechanism of DQN and graph mining can provide effective evaluation and prediction for the mental health status of college students.

In the context of the “Internet plus” era, the spread and exchange of catering culture is an indispensable part of people’s daily life. The historical catering culture evolution system is based on geographic information system (GIS) technology, which uses historical literature as the basic data source and combines computer graphics, spatial analysis and other technical means to achieve visual expression of the historical catering culture evolution process. This system visualizes the spatiotemporal evolution of historical catering culture, helping people better understand the characteristics of catering culture in different eras and regions. Based on this, the reasons for the formation of differences in catering culture in different regions are explored, providing a foundation for further in-depth research on historical catering culture. The results showed that the data accuracy of the GIS-based historical catering culture evolution system reached the highest of 96.1%. At the same time, performance efficiency and functional integrity have also been improved.

With the development of Internet technology, the application of big data and artificial intelligence (AI) can be seen more and more. Therefore, in modern society, people are no longer satisfied with simple learning and research, but are increasingly pursuing a more in-depth and comprehensive exploration. The talent competency model prediction platform is an important tool in current talent selection and management. However, current research has shortcomings in terms of prediction accuracy and efficiency. This article introduced a talent competency model prediction platform based on intelligent interactive personality testing software and intelligent clustering analysis. This platform utilized advanced technological means to analyze individual personality traits and perform intelligent clustering to predict their competence in the workplace. Research has shown that this method can significantly improve the efficiency and accuracy of talent selection and development, providing decision-making support for enterprises. The average efficiency was 93.92%, and the average accuracy reached 93.4%. In addition, it can significantly reduce costs.

With the acceleration of globalization, the strategy of sustainable development has been widely recognized by the international community. Financing issues have also been receiving attention from all parties, and debt financing, as the main financing method of enterprises, has been the target and direction that enterprises are competing for. This study employs data from China’s A-share heavily polluting listed companies between 2013 and 2022 to investigate the influence of debt financing on the level of environmental information disclosure. The findings reveal that debt financing positively impacts the level of corporate environmental information disclosure. Specifically, as corporate debt rises, businesses tend to enhance their disclosure levels by policy and market demands. Notably, commercial credit borrowing serves as a reflection of enterprises’ market behavior.

With the continuous development of computer network technology, the traditional manufacturing mode is only suitable for small workshop mode, and there is also the problem of information island, which is difficult to meet the needs of current intelligent factory system construction. This paper explores the application of artificial intelligence technology in mechanical manufacturing, and constructs an artificial intelligence model through practical cases. In order to improve the production management efficiency of enterprises and reduce production costs, manufacturing enterprises integrate advanced technologies such as information technology, automation technology, modern management technology and system science and technology into traditional manufacturing technology. The DNC and MES integrated system developed in this paper uses Ethernet technology and TCP/IP protocol to realize the communication between the system and field devices. Moreover, this study builds the verification environment based on the hardware environment of Window system workstation in the laboratory. By analyzing the experimental data, we can see that the artificial intelligence mechanical manufacturing simulation system proposed in this paper can effectively improve the efficiency and quality of mechanical manufacturing.

With the global sustainable development goal and the rise of corporate social responsibility, ESG (environmental, social and governance) information transparency has become an indispensable part of enterprise management. The purpose of this paper is to explore how blockchain technology can be used as an innovative tool to provide effective support for improving the transparency of ESG information. Firstly, this paper establishes a theoretical framework, drawing lessons from transaction cost economics, information asymmetry theory and corporate governance theory to understand the theoretical impact of blockchain technology on ESG. Subsequently, through empirical research, this paper collects and analyzes ESG information of enterprises in different industries and regions, and shows the actual effect of blockchain technology in ESG compliance, data accuracy and transparency. The research results show that blockchain technology has significantly improved the overall performance of enterprises in the ESG field. However, while achieving results, this paper also found room for improvement in standardization, technical performance and smart contracts. Finally, this paper puts forward some suggestions for future research, including in-depth industry research, long-term impact research and investigation of regulations and policies, in order to provide more reliable theoretical and empirical support for the wide application of blockchain technology in ESG information transparency. This study provides far-reaching enlightenment for building a more transparent and sustainable business ecosystem.

This study aims to explore customer sentiment and demand prediction models based on cognitive computing to improve companies’ understanding of customer behavior and optimize market strategies. By combining deep learning and natural language processing technology, this paper builds a sentiment analysis model that can effectively extract emotional tendencies from customer feedback, and a demand prediction model that combines sentiment analysis results and historical transaction data to predict future customer needs. Empirical analysis shows that the sentiment analysis model exhibits high accuracy, while the demand forecast model shows good predictive performance, proving the effectiveness of the comprehensive model in customer demand forecasting. In addition, this study also discusses the application potential of the model, providing companies with a new tool to better understand customer behavior and develop precise market strategies. Although there are certain limitations, such as the impact of the size and diversity of the data set on the generalization ability of the model, and the need to improve the interpretability of the model, this study provides a basis for future research in this field and points out possible research directions, including expanding data sources, exploring new data analysis techniques and algorithms, and enhancing the versatility and adaptability of models.