課程教學(xué)大綱英文版商務(wù)數(shù)據(jù)分析

1、Syllabus of Business data analysisCourse Name： Business data analysisCourse Code：Credits： 3.0Total Credit Hours： 48Lecture Hours： 40Experiment Hours： 0Programming Hours： 8Practice Hours： 0Total Number of Experimental (Programming) Projects 4 ,Where, Compulsory ( 4 ), Optional ( 0 ).School ： School o

2、f BusinessTarget Major： Business managementI、Course Nature & AimsBusiness data analysis11 is a professional elective curriculum for majors in economics and management. Through the teaching of this course, students can master the basic principles of deep learning, convolutional neural networks, recur

3、rent neural networks, generative adversarial networks, attention mechanisms and other basic methods and their typical application fields, and use machine learning open source platform TensorFlow to achieve deep learning applied in many typical fields such as securities forecasting, sound quality eva

4、luation, electronic recommendation, target detection, and social network sentiment analysis, the course plays an important role as a link between theory and practice. Pay attention to the combination of qualitative analysis and quantitative analysis, strengthen practical training, and lay a good fou

5、ndation for students* subsequent learning, practice and future work and development.II Course Objectives1. Moral Education and Character Cultivation.Learn business data analysis theory and technology through the course, have a comprehensive understanding of deep learning related knowledge. Through t

6、he explanation of the history of business data analysis and the establishment of relevant theories and technologies, understand how predecessors think in the process of business data analysis development, how to overcome the obstacles encountered, help students establish the scientific thinking meth

7、ods and the spirit that dare to face the challenges in their work. From the perspective of the role of business data analysis disciplines in Chinas innovation-driven development, with the research work of outstanding contributors as the carrier, the education of socialist core values is integrated i

8、nto the course content and the entire process of the teaching process, highlighting value guidance, knowledge transfer and ability training to help animal recognition as an example to understand the application of the VGG convolutional neural network model, improve the ability to analyze and solve p

9、ractical problems using related convolutional neural networks.Key Points： 4.1 4.2、4.3、4.4Difficult Points： 4.1、4.4Recurrent neural network (supporting course objectives 2)Basic principles of recurrent neural networksRecurrent neural network modelLong and short-term memory neural network modelStock p

10、rediction based on LSTMTeaching Requirements: Through the study of this chapter, students are required to understand the basic principles of recurrent neural networks, be familiar with the structural characteristics of long and short-term memory models, take stock prediction as an example to underst

11、and the application of LSTM, and have the ability to use LSTM to analyze and solve practical problems.Key Points： 5.2 5.3、5.4Difficult Points： 5.2、5.3Target detection (supporting course objectives 2)Overview of target detectionBasic concepts of target detectionDevelopment of target detectionTarget d

12、etection based on candidate regionsFast R-CNN target detection algorithmFaster R-CNN target detection algorithmYolo target detection algorithmCase analysis of target detectionComplete experimental (programming) project 4: Drawing of histogram and scatter diagramTeaching Requirements: Through the stu

13、dy of this chapter, students are required to understand the overview of target detection, understand the basic concepts of target detection, master Fast R-CNN target detection algorithm, Faster R-CNN target detection algorithm, Yolo target detection algorithm, and through the target detection case a

14、nalysis, familiar with the application of target detection.Key Points： 6.4、6.5、6.6 6.7、6.8Difficult Points： 6.5、6.6 6.7Generative adversarial network (supporting course objectives 2)Basic principles of generating adversarial networksEncoder-Decoder modelGenerating an adversarial network algorithm DC

15、GANGenerating Adversarial Network Algorithm ApplicationHandwriting generationTeaching Requirements: Through the study of this chapter, students are required to understand the basic principles of generating adversarial networks, familiar with the Encoder-Decoder model, the generation of adversarial n

16、etwork algorithms DCGAN, and the use of handwriting generation as an example to master the application of generating adversarial networks, have the ability to use generative confrontation network to analyze and solve practical problems.Key Points： 7.2 7.3、7.5Difficult Points： 7.3Attention mechanism

17、(supporting course objectives 2)Seq2Seq modelAttention mechanism modelMachine translationTeaching Requirements: Through the study in this chapter, students are required to understand the Seq2Seq model and the attention mechanism model, take machine translation as an example to master the application

18、 of the attention mechanism, and have the ability to use the attention mechanism to analyze and solve practical problems.Key Points： 8.2 8.3Difficult Points： 8.2Deep learning applications (supporting course objectives 3)Sound quality evaluationCodecEmotion recognition classificationComplete experime

19、ntal (programming) project 1: Text classificationComplete experimental (programming) project 2: Sentiment analysisComplete experimental (programming) project 3: Target detectionComplete experimental (programming) project 4: Machine translationTeaching Requirements: Through the study in this chapter,

20、 students are required to understand audio samples and feature preprocessing, audio feature selection, LeNet convolutional neural network model training in audio quality evaluation, familiar with codecs, master emotion recognition classification, and further improve the ability to use deep learning

21、theoretical methods to analyze and solve practical problems.Key Points： 9.1 9.3Difflcult Points： 9.3IV、Table of Credit Hour DistributionTeachingContentIdeological and Political IntegratedLectur e HoursExperiment HoursPractice HoursProgramming HoursSelf-study HoursExercise ClassDiscussion HoursNeural

22、 network basicsCorrectvalues4000000The application of deep learning in artificial intelligenceBasic national conditions, historical laws4000000CNN convolutional neural networkScientific methodolog y6000000Typical convolutional neural network algorithmStrict scientific attitude4000000Recurrent neural

23、 networkScientificmethodolog4000000yTarget detectionCorrectvalues6000000Generative adversarial networkStrict scientific attitude4000000AttentionmechanismScientific methodolog y4000000Deep learning applicationsCorrectvalues4000000Experimental (programming )project 1: Text classificationScientificmeth

24、od0002000Experimental (programming)project 2: Sentiment analysisScientificmethod0002000Experimental (programming )project 3: Target detectionScientificmethod0002000Experimental(programming )project 4: Machine translationScientificmethod0002000Total40008000Sum48V、Summary of Experimental (Programming)

25、 ProjectsNo.Experiment(Programming)NameExperiment(Programming)HoursExperiment(Programming)TypeAbstractCourseTypeSupportingCourseObjectives1Experimental(programming) project 1:Text classification2comprehensiveText classificationusing convolutionalneural networkCompulsory32Experimental(programming) pr

26、oject 2:Sentiment analysis2comprehensiveSentiment analysisusing LSTMCompulsory33Experimental (programming) project 3: Target detection2comprehensiveTarget detection usingYolo target detectionalgorithmCompulsory34Experimental (programming) project 4: Machine translation2comprehensiveMachine translati

27、on using attention mechanismCompulsory3Note: Fill in ncomprehensive, designing, verification and demonstration in the experiment type, which refers to comprehensive experiment, designing experiment, verification experiment and demonstration experiment, respectively.Comprehensive experiment refers to

28、 the experiment involving the comprehensive knowledge of the course or related knowledge of the course.Designing experiment refers to the experiment in which the students design the experiment scheme and complete it by themselves based on the given experimental objectives and experimental conditions

29、.Verification experiment refers to an experiment conducted to verify whether the knowledge or hypothesis is correct after a certain understanding of the research object is achieved or a certain hypothesis is put forward.Demonstration experiment refers to the experiment performed by the teacher in li

30、ne with the teaching content.VI、Teaching MethodThis course is mainly taught by teachers in class, supplemented by self-study based on course video materials and homework. The teaching process should be able to flexibly use blackboard writingand multimedia teaching, strengthen teacher-student interac

31、tion, pay attention to heuristic teaching, for the important knowledge points, use the ideas of putting forward questions, analyzing problems and solving problems to teach, subtly cultivate students1 corresponding ability.VD、Course Assessment and Achievement EvaluationAssessment Methods： Non-Examina

32、tionExamination Formats： Open-bookGrading Methods： Five-level SystemCourse Assessment Content, Assessment Format and Supporting Course ObjectivesCourseObjectives (Indices)AssessmentContentAssessment Formats and Proportion ( % )GradingClassroomQuestioningAssignmentEvaluationRoutineTestExperimentRepor

33、tTermReportTermPaperMidtermExamFinalExamProportion(%)Objective 1. Master the relevant knowledge of deep learning basic principles and development trends.(Index 2 6、8、9、10、11)Definition of neural network, activation function, loss function, learning rate, overfitting, development of neural network.25

34、000001017Objective 2. Master the knowledge of basic methods of deep learning.（Index 1、2、 3、 4）Convolution operation, pooling operation, LeNet5 convolutional neural network, AlexNet convolutional neural network model, VGG convolutional neural network model, GoogLeNet convolutional neural network mode

35、l, ResNet convolutional neural network model, basic principles of recurrent neural network model, long Short-term memory neural network model, Fast R-CNN target detection algorithm, Faster R-CNN target detection algorithm, Yolo target detection algorithm, generated adversarial network620000004066alg

36、orithm DCGAN, Seq2Seq model, attention mechanism model.Objective 3. Master the skills of using deep learning related knowledge to solve practical problems. (Index 1 2、 3、 4、 5、7)The specific application of deep learning theoretical methods.25000001017Total10300000060100VID、Course ResourcesTextbooks：

37、Zhao Weidong, Dong Liang. Machine learning M. Post and Telecom Press, 2018.Bibliography：Zhao Weidong. Machine learning case combat fM. Post and Telecom Press, 2019.Zhao Weidong, Dong Liang. Python machine learning combat case M. Tsinghua University Press, 2019.Reading Materials：CNKI China HowNet, SD

38、OL (ScienceDirect Online) and other digital resource libraries include academic papers related to deep learning.Newspapers, magazines, books and other related materials in college collections.Academic website materials related to deep learning at home and abroad.IX、NotesPrerequisites: Machine learni

39、ngFollow-up Courses: NoContents and Requirements of Students1 Self-study: NoBilingual Teaching or Not: NotRequirements and Proportion of Bilingual Teaching: NoDiscipline and Considerations of Practice Session: no practice sessionNotes: Nostudents correctly understand the laws of history, accurately

40、grasp the basic national conditions, grasp the scientific world outlook, methodology, and promote the establishment of a correct world outlook and values.Course ObjectivesThrough the study of this course, students qualities, skills, knowledge and abilities obtained are as follows：Objective 1. Master

41、 the relevant knowledge of deep learning basic principles and development trends. (Corresponding to Chapter 1 2, supporting for graduation requirements index 1-1 l-2 l-3 3-1、 3-2 33、 7-k 7-2、7-3、8-1、8-2、8-3、9-1、9-2、9-3、10-1 10-2、103 11-k 11-2、11-3、12-1、12-2、12-3)Objective 2. Master the knowledge of

42、basic methods of deep learning. (Corresponding to Chapter 3 4、5、 6、 7、 8, supporting for graduation requirements index 2-1 2-2、 2-3 3-1 3-2 3-3、 41、 4-2、 4-3、 5-1 5-2、5-3)Objective 3. Master the skills of using deep learning related knowledge to solve practical problems. (Corresponding to Chapter 9

43、, supporting for graduation requirements index 21、 22、 2-3 31、 32、 33、 4-1、 4-2、4-3、5-1、5-2、5-3、6-1、6-2、6-3、8-1、82 8-3). Supporting for Graduation RequirementsThe graduation requirements supported by course objectives are mainly reflected in the graduation requirements indices 1-1 l-2 l-3 21、2-2 2-3

44、、3-1、3-2 33、4-1、4-2、4-3、5-1 52、5-3、6-1、 6-2. 6-3 7-1、7-2、7-3、8-1 8-2、8-3 9-1、92、9-3 10-1、10-2. 10-3 ll-K ll-2 63 12-1 122、 12-3, as follows:Supporting for Graduation Requirementsprofessional ethics.CourseObjectivesGraduationRequirementsIndices and Contents Supporting for GraduationRequirementsTeachi

45、ngTopicsLevel ofSupportIndicesContentsObjective 1Master the relevant knowledge of deep learning basic principles and development trends1-K 12、13、31、3-2 3-3 7-1 7-2、7-3、 8-1、 8-2、 8-3、9-1、9-2、9-3、10-1、10-2、10-3、11-1、11-2、11-3、12-1、122 12-31-1 Love the socialist motherland, support the leadership of t

46、he Communist Party of China, have a firm and correct political direction. 1-2 Have a correct world outlook, outlook on life and values, and a sense of home country and craftsmanship spirit.1-3 Have good ideological and moral character, social morality andChapter 1、2L3-1 For the professional foreign

47、language, computer and other tools and methods involved in industrial engineering.3-2 Methods and skills in modern industrial engineering, system optimization, management prediction and decision-making, etc.3-3 Have strong self-study ability, innovative spirit and entrepreneurial consciousness.7-1 H

48、ave good professional spirit and professional altitude.7-2 Have the ideal, ambition and sense of mission to strive for the prosperity of the country and the people and the rejuvenation of the nation.7-3 Have a high sense of social responsibility, integrity awareness, abide by professional ethics and

49、 norms.8-1 Be able to effectively communicate with peers in the industry and the public on complex engineering issues.8-2 Write reports and design manuscripts, make statements, express clearly or respond to instructions.8-3 Ability to communicate in across-cultural context.9-1 Have strong self-study

50、 ability, innovative spirit and entrepreneurial consciousness.9-2 With the concept of lifelong learning, with the spirit of exploration and ability.9-3 Have creative thinking and ability to implement.10-1 Understand the frontiers of professional development and social needs.10-2 Have the ability to

51、adapt to the future technology development and management change.10-3 Have the ability to deal with uncertainty.11-1 Professional knowledge of engineering, management and economics.11-2 Knowledge of mechanical manufacturing, industrial design, intelligent manufacturing and other fields.11-3 Be able

52、to use interdisciplinary knowledge and thinking to analyze and solve problems.12-1 Have the humanities knowledge of history, art, philosophy, etc.12-2 Understanding humanistic thoughts and methods.12-3 Follow the humanistic spirit.Objective 2Master the knowledge of basic2-1、2-2、2-3、3-1、3-2、3-3、4-1、4

53、-2、2-1 Have solid knowledge of mathematics and natural science.Chapter 3、4、5、6、7、Mmethods of deeplearningmethods of deeplearning4-3、5-K 5-2、5-32-2 Master the basic theories, methods and tools of engineering technology systematically.2-3 All kinds of engineering knowledge can be used to solve enginee

54、ring comprehensive problems.3-1 For the professional foreign language, computer and other tools and methods involved in industrial engineering.3-2 Methods and skills in modern industrial engineering, system optimization, management prediction and decision-making, etc.3-3 Have strong self-study abili

55、ty, innovative spirit and entrepreneurial consciousness.4-1 Be able to apply the basic principles and theories of natural science and management science to analyze and discover the efficiency, quality, cost and environmental optimization of production and service system.4-2 It can design and realize

56、 the optimization method of efficiency, quality, cost and environment of production and service system based on practical management problems4-3 Be able to embody the sense of innovation in the design process,and consider the factors of society, health, safety, law, culture and environment.5-1 Basic

57、 ability of data acquisition, processing and analysis.5-2 It can adapt to the needs of digital transformation, and has data thinking and the ability of data processing.5-3 Ability of data modeling, result analysis and interpretation.Objective 3Master the skills of using deep learning related knowled

58、ge to solve practical problems2-1、22、23 3-1、32、33、4-1、42、4-3、5-1、5-2、5-3、6-1、6-2、6-3、8-1、8-2、8-32-1 Have solid knowledge of mathematics and natural science. 2-2 Master the basic theories, methods and tools of engineering technology systematically.2-3 All kinds of engineering knowledge can be used to

59、 solve engineering comprehensive problems.3-1 For the professional foreign language, computer and other tools and methods involved in industrial engineering.3-2 Methods and skills in modern industrial engineering, system optimization, management prediction and decision-making, etc.3-3 Have strong se

60、lf-study ability, innovative spirit and entrepreneurial consciousness.4-1 Be able to apply the basic principles and theories of naturalChapter 9Mscience and management science to analyze and discover the efficiency, quality, cost and environmental optimization of production and service system.4-2 It