Software cost estimation is an important task in software project development and management. Based on the research results of renowned economist Boehm, this article constructs the COCOMOII model, which summarizes software cost estimation into five steps: workload estimation, proportion factor estimation, workload multiplier estimation, development progress estimation, and currency cost estimation, and presents them in a flowchart. Based on the COCOMOII model, this paper takes the “Land Sea Intermodal Logistics Information Platform” developed by the author for a certain company as an example to illustrate the process and method of software project cost estimation based on the COCOMOII model. In practical application, it is necessary to fully absorb the advantages of other estimation methods, enhance the flexibility, scalability, and operability of the model, in order to demonstrate strong vitality and make software development cost estimation closer to the actual value.

Aiming at the challenges such as the difficulty in identifying new variant viruses in computer virus detection, the slow convergence speed of traditional algorithms, and the limited edge computing resources, a new virus detection paradigm integrating quantum computing and improved CSA is proposed. Firstly, the global search ability of the algorithm is enhanced through dynamic affinity calculation and a two-stage clone amplification strategy. The quantum superposition state mutation operator and the quantum tunneling effect adaptive mutation rate adjustment mechanism are introduced to break through the local optimal bottleneck of the classical algorithm in the high-dimensional solution space. Secondly, the storage structure of the immune memory bank is optimized by utilizing the characteristics of quantum entanglement to achieve parallel retrieval and distributed update of the historical optimal solution, reducing the memory occupation on edge devices to 1/5 of the traditional method. Experiments show that although CAS is slightly inferior to the CNN algorithm in terms of detection accuracy rate, false alarm rate and missed alarm rate, it has certain advantages in detection speed and is significantly better than FCM. The research results have broken through the limitations of traditional detection technologies and achieved effective detection of unknown viruses.

Knowledge bases are important collections of knowledge that store, organize, and process knowledge, as well as provide knowledge services. Knowledge bases are an important part of building an open infrastructure that collects and preserves large amounts of research content and provides access to it. The study investigates the key technologies for building the Japanese sentence pattern knowledge base based on Oracle database. First, the storage structure of Oracle database is studied; then, the knowledge acquisition based on web crawler technology is studied; then, the requirement analysis is conducted based on the use case diagram technology; finally, the conceptual structure design and logical structure design of the knowledge base storage structure are conducted. The research results of this paper, which conform to the basic specifications of software engineering, provide a complete solution for the construction of Japanese sentence patterns knowledge base and lay the foundation for software development and implementation.

Network security covers a wide range, mainly to ensure the integrity and effectiveness of network theory and network technology information, fully grasp the security state of the entire network system, has become an urgent need for managers and decision makers. This paper constructs an evaluation index system composed of “physical and environmental security, network and communication security, software and data security, network security precautions, personnel and management security, network security training”. In this paper, a hierarchical cluster analysis model is constructed, which is composed of “constructing initial sample matrix, sample matrix data standardization, deviation square and clustering strategy”, and its scientificity is verified by experiments. The experimental data of 24 representative network systems were selected, and the experimental results were analyzed by means of case processing summary, clustering process, icicle chart and clustering pedigree. The results show that the mathematical model proposed in this paper can scientifically classify network systems and provide basic support for network security management.

Make a feasible schedule planning for software project, is the foundation to carry out an orderly software project activity, is the key to success of the project. To solve the difficulties of schedule planning, this article studies the science technology and methods, including four aspects: First, using network chart to show the dependencies between activities, including Precedence Diagram Method and Arrow Diagram Method; Second, Program Evaluation and Review Technique is used to evaluate the time of the project activity; Third, the critical path method of determining the project total duration, including deduce from calculation formula of active time and determine the critical path; Fourth, duration compression method based on progress compression factor method. According to the situation of many technologies and methods, studied the main factors considered by selecting techniques and methods. Results show that using scientific technology and method to make the schedule planning, will reduce the workload of planning, improve the accuracy of the plan and have high practical value.

The network teaching system provides an opportunity for new educational methods and expands educational functions in an all-round way. Network teaching breaks the traditional classroom teaching mode, breaks through the effective communication and exchange that cannot be implemented in distance education. Database design is an important work in software development. In the process of network teaching system development, from the perspective of user needs, the conceptual structure design and logical structure design of the database are carried out according to the standardized method. The result of conceptual structure design is represented by E-R diagram, which is composed of 5 entities and 5 links between them; the logical structure design is based on Oracle database management system, and the conceptual structure model is transformed into a specific table structure. Improve data storage efficiency and access speed by optimizing design results.

The Python language is more focused on problem solving, which is in line with the era of computational thinking, and the teaching of Python language course requires students to systematically master the basic concepts, programming ideas and programming techniques of Python, and have the idea of object-oriented software design technology. We have developed a recommended system of online resources for "Python Programming" course to solve the problem of students' access to resources and deepen the teaching reform of the Python Programming course. Data persistence is an important task in system development, and Hibernate is the most popular O/R Mapping framework. The data persistence design based on Hibernate solves the key technical problems in the development of the network resource recommendation system for the "Python Programming" course, and improves the efficiency and maintainability of the software system development.

System evaluation is an important branch of system science and the main content of the system engineering, and it also provides a scientific basis for the system evaluation. This paper presents a comprehensive evaluation model based on nonlinear classify algorithm of SVM. First of all, study and evaluate the foundation work, including evaluation index system, the index weight determination and the treatment of data standardization; Then, the research SVM, accurate description of nonlinear classify algorithm after clarify the basic thought, shows the five kinds of commonly used kernel function, and improve traditional SVM aiming at the existing problems. The content of this paper effectively solves the problem about model selection and over learning, nonlinear and the dimensionality curse, local minimum, it also improves the training speed and precision of SVM at the same time, reducing the amount of calculation, and it has very strong adaptability and practical value.

BP algorithm has the advantages of simple structure, easy to realize, and it has been widely applied in fault diagnosis, pattern recognition, but the shortcoming of BP algorithm has affected its further development. Based on constructing the mathematical model of the BP algorithm, this paper analyzes the shortcomings of BP algorithm. Problems exist in the slow learning speed, the large possibility of failure to learn, poor generalization ability, the presence of multiple extreme points, the network structure is not easy to determine. Based on research results, heuristic information technology and algorithms, numerical optimization methods, this paper focuses on the study of BP learning algorithm and several improved methods, including improve the algorithm on convergence rate of the variable learning rate and algorithm adding momentum item, improve the search ability and generalization ability of genetic algorithm, eliminate resilient BP algorithm of Gradient magnitude effect, overcome the simulated annealing algorithm of the local minimum. The improvements of BP learning algorithm have a very important significance, especially when combined with other intelligent algorithms, whether in academic or in the application.

Public opinion analysis under the big data environment pays more attention to the collection, storage and cleaning of Internet data, and combines text mining technology to obtain opinion information from a large amount of low-value data. Based on the Hadoop platform, this paper develops the Internet opinion analysis system for emergencies, timely predicts and intervenes in the development trend of Internet opinion, effectively responds to negative opinions, and improves the level of social governance. First, the Hadoop ecosystem architecture consisting of “low-level components, data storage, distributed computing, data analysis and processing, and result output” was designed; then, the JDBC database access and optimization technology was designed; finally, it was designed by data system functions such as “data collection, data storage, data processing, data analysis, and data display”.

Mobile App has the characteristics of good interactivity, clear logic, simple operation and fast response speed, providing users with rich experience. In the era of mobile information, college students carry their mobile phones with them, and it is the easiest and fastest way to use mobile apps to check their scores. Aiming at the high market share of iPad and iPhone among college students, this paper refers to the design and business process of similar domestic apps, and sorts out the functional modules of the app based on demand analysis, and designs a fully functional and easy-to-operate student score query app, and it is convenient for students to check their scores anytime and anywhere. The core content of the research is based on the hierarchical structure of the iOS platform, which completes the design of the score query process and the design of data access, and provides the complete solution for App development.

Data pre-processing is to process the collected original data and prepare for data analysis. In this paper, based on MapReduce technology, MapReduce calculation is effectively decomposed into Map and Reduce calculation process to maximize the degree of parallelism. The framework and workflow of MapReduce are analysed. The core work consists of three parts. One is text pre-processing, which divides the text data into several data blocks with appropriate size, and then processes them by Map function and Reduce function respectively. Second, feature selection, using Bayesian classifier, is divided into preparation stage, training stage and application stage. The third is text vectorization, which uses TF-IDF algorithm to represent text as a series of vectors that can express text semantics.

城市突发事件应急管理的效果关系到社会稳定、城市发展和人民生活幸福,大数据技术为政府应急管理提供了全新的解决方案.文章基于大数据思维和创新驱动发展的理念,分析了大数据在政府应急管理中的应用价值,指出了大数据时代政府应急管理面临的困境,提出了大数据时代城市突发事件政府应急管理创新举措.将以大数据技术为代表的新信息技术应用于应急管理,提高数据利用效率及完善基于数据支撑的应急管理体制建设,是大数据时代政府应急管理变革的必由之路.

MOOC is the application of information technology in the field of education, changing the traditional teaching mode and providing students with a free and open learning environment. Based on the relevance learning theory, humanistic learning theory and educational psychology theory, this paper points out the advantages of MOOC in learning time, solving passive learning problems, learning resources more targeted, and strengthening interaction and communication. This paper analyzes the problems of fuzzy training objectives, less total class hours, and different computer foundations in computer basic course teaching. It also proposes to classify teaching resources, speed up teaching methods and means innovation, strengthen classroom interactive learning design, rationally transform teachers and students, and adopt a multi-evaluation system and other MOOC-based computer basic course teaching reform measures.

The emergence of big data technology has completely changed people's understanding of natural and social thoughts, views and accuracy. Many past research hypotheses, management models and methods have been re-recognized or even overturned. Guided by the related theories of contingency management and big data, this paper points out the basic connotation of big data, analyzes the impact of innovation management and big data on social governance environment in the era of big data, and constructs the basic application of big data in contingency management. The framework presents new challenges for government contingency management in the era of big data. The wide application of big data technology in the field of contingency management will effectively enhance the data capacity of China's contingency management system, and even reshape the system, mechanism and process of contingency management.

Establishment of the performance appraisal Index system of software project pre-sales manager is are important in human resources management work of the software enterprises, software project can sign the contract, depend on the software project manager's job performance. Performance appraisal for the current problems, use of balance scorecard and critical success factors method, Focus on performance appraisal index system and method to study the process. First analyzes the idea of setting up performance goals, then clear the strategic goals into the organization method of measure, finally, use software project pre-sales manager processes and control points with performance appraisal index system, establish the key performance appraisal index. The result show that performance appraisal is a complex system engineering, balanced score card method and the combination of key success factors is the establishment of pre-sales manager, software project performance appraisal index system of effective methods and effective performance appraisal contribute to corporate strategic objectives achieve.

SSH is a powerful open source framework of Java Web and a lightweight solution for develop J2EE enterprise-class Web application. For the characteristics of SSH framework configuration file content more, configuration complexity and difficult to grasp, study the core configuration files. Based on the study of web.xml file that integrated MVC framework and Web applications, research the configuration techniques and methods such as the core configuration file struts.xml of Struts framework, the core configuration file Context.xml of Spring Framework and the core configuration file hibernate. cfg.xml of Hibernate framework. This study enables developers to clearly understand the way to write the core configuration file of SSH framework, and contribute to team members work in parallel and improves development efficiency.

In the highly competitive market environment, after-sales service has become the key for enterprises' survival.In order to solve the problem of difficult management of after-sales service, this paper surrounds framework improvements, system architecture workflows and business functions to construct the after-sales service system which is based on SSI framework.First of all, framework improvement processes are detailed introduced, improved methods on each layer are detailed elaborated; then, based on the improved framework, the system architecture workflow is implemented; finally, main business functions implemented by the system are described.The implementation of the system can centrally manage the repair, maintenance, installation and other daily work; standard after-sales service business processes, information manage spare parts, reduce the manpower, material resources and costs, which greatly improves the quality and management efficiency of after-sales service.

XML has been widely studied as the standard of data exchange in e-government applications. With the arrival of the era of big data,the management of XML data in e-government is also becoming more and more important. In the management of XML data,the similarity of XML documents is the key of XML data integration and XML data classification. In order to study the XML document simi-larity,the XML document are transformed into tree,extracting the corresponding characteristics of the nodes of the tree,and then using these characteristics to calculate the similarity of nodes,and then the final node similarity can be obtained by the ELM( Extreme Learning Machine) algorithm. Based on the similarity of nodes,the algorithm of similarity comparison of the XML document tree is given,which can obtain the similarity of XML documents. Based on the given specific evaluation indexes,the accuracy,recall, F-measure values and the corresponding time are obtained through experiments in two different data sets using the method proposed. The performance advanta-ges of the proposed method are verified by experiments.

Pragmatic Failure in Intercultural Communication is a hot issue in recent years, with language learning and intercultural communication studies, Pragmatic manifestation of failure is an important part of your mistakes in the field of research. Firstly, to build cross-cultural Pragmatic Communication Failure manifestations structure, then we studied the language pragmatic failure, social pragmatic failure and pragmatic behavior failure. Wherein, language pragmatic failure, pragmatic person research capacity due to the lack of proper understanding, or lack of use of language features to express the intention of generating errors; Social pragmatic failure, of different cultural backgrounds language communicators language habits and characteristics, so decent use of language and avoid mistakes; Pragmatic behavior failure, research is needed to transmit information through the use of methods or means do not belong to the category of speech, expression expressive approach.