7. 附錄
7.3. 一般文字稿 Chapter 數作為主題數
未刪章節未加入提示詞時各摘要方法與 PLSA 摘要方法和一般 LDA 摘要方 法比較結果如表 64 所示,與動詞 LDA 摘要方法和名詞 LDA 摘要方法比較結果 如表 65 所示,與線上摘要器 SweSum 比較結果如表 66 所示。
表 64. 與 PLSA 摘要方法和一般 LDA 摘要方法比較結果
課程名稱 一般
LDA
TF TF-ISF TF-SF PLSA The Hardware/Software Interface 0.3126 0.3245 0.3244 0.3191 0.3079 Social Network Analysis 0.3405 0.3578 0.3588 0.3535 0.3296 Web Application Architectures 0.4236 0.4539 0.4476 0.4532 0.4304 Audio Signal Processing for
Music Applications
0.1734 0.1897 0.201 0.1897 0.1791 Malicious Software and its 0.3572 0.3629 0.3585 0.3633 0.3409
146 Underground Economy Two Sides
to Every Story
Big Data Science with the BD2K-LINCS Data Coordination and Integration Center
0.3295 0.3398 0.3393 0.3389 0.331
Experimental Methods in Systems Biology
0.3241 0.3303 0.3354 0.3322 0.3332 Dynamical Modeling Methods for
Systems Biology
0.3341 0.3513 0.3517 0.342 0.3347 The Brain and Space 0.2833 0.3074 0.3026 0.3041 0.2978 Network Analysis in Systems
Biology
0.3122 0.3302 0.3283 0.3254 0.3227
automata 0.5121 0.52 0.523 0.5178 0.5083
Natural Language Processing 2013
0.2717 0.2813 0.2742 0.2791 0.2808 Beginning Game Programming
with C
0.2788 0.3326 0.3325 0.3214 0.2765 Climate Change 0.2987 0.3166 0.3203 0.3241 0.2851 Journalism Skills for Engaged
Citizens
0.3474 0.353 0.3515 0.3534 0.351 Algorithms Design and Analysis
Part 1
0.284 0.3135 0.3123 0.315 0.2876 Algorithms Design and Analysis
Part 2
0.3198 0.3439 0.3421 0.3406 0.3265 Introduction to Chemistry
Reactions and Ratios
0.2704 0.2856 0.2902 0.2859 0.2726 Genomic and Precision Medicine 0.2874 0.2854 0.2897 0.284 0.2866 Epigenetic Control of Gene Surviving Disruptive
Technologies
0.4158 0.4243 0.4186 0.423 0.4041 Caries Management by Risk
Assessment CAMBRA
0.3657 0.4094 0.4214 0.3989 0.3429 Computational Neuroscience 0.3143 0.3186 0.3191 0.32 0.3084 Introduction to Data Science 0.2598 0.2675 0.266 0.2669 0.2571 Discrete Optimization 0.2928 0.3003 0.3028 0.2973 0.2846 Foundations of Virtual Instruction 0.2769 0.2976 0.2856 0.2864 0.232 Virology I How Viruses Work 0.2429 0.2495 0.2501 0.2432 0.238 Edx Introduction to Computer
Science edx Introduction to Programming
with Java Part 1 Starting to Code with Java
0.1292 0.1415 0.1394 0.1384 0.134
edx Programming Basics 0.3789 0.3738 0.3752 0.3714 0.3789 總平均 0.3155 0.3297 0.3292 0.3276 0.3126
平均 F1 值:與一般 LDA 摘要方法呈現顯著差異
平均 F1 值:與 PLSA 摘要方法呈現顯著差異
147
紅色字體平均 F1 值:平均 F1 值未比一般 LDA 摘要方法或 PLSA 摘要方法高
在表 64 的結果中除了 edx Cellular mechanisms of brain function 課程和 edx Programming Basics 課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平 均 F1 值皆比一般 LDA 摘要方法或 PLSA 摘要方法低,其餘課程的 TF 摘要方法、
The Hardware/SoftwareInterface
0.3158 0.3245 0.3244 0.3191 0.313 Social Network Analysis 0.3466 0.3578 0.3588 0.3535 0.3503 Web Application Architectures 0.4286 0.4539 0.4476 0.4532 0.4399 Audio Signal Processing for
Music Applications
0.1879 0.1897 0.201 0.1897 0.1734 Malicious Software and its
Underground Economy Two Sides to Every Story
0.3101 0.3629 0.3585 0.3633 0.31
Big Data Science with the
BD2K-LINCS Data
Coordination and Integration Center
0.3408 0.3398 0.3393 0.3389 0.3425
Experimental Methods in Systems Biology
0.3268 0.3303 0.3354 0.3322 0.3313 Dynamical Modeling Methods
for Systems Biology
0.337 0.3513 0.3517 0.342 0.3476 The Brain and Space 0.2957 0.3074 0.3026 0.3041 0.2933 Network Analysis in Systems
Biology
0.3174 0.3302 0.3283 0.3254 0.3173
automata 0.5086 0.52 0.523 0.5178 0.5134
Natural Language Processing 2013
0.2711 0.2813 0.2742 0.2791 0 Beginning Game Programming
with C
0.2924 0.3326 0.3325 0.3214 0.2988 Climate Change 0.312 0.3166 0.3203 0.3241 0.3028 Journalism Skills for Engaged
Citizens
0.3484 0.353 0.3515 0.3534 0.3554 Algorithms Design and Analysis
Part 1
0.2863 0.3135 0.3123 0.315 0.2879 Algorithms Design and Analysis
Part 2
0.32 0.3439 0.3421 0.3406 0.3166 Introduction to Chemistry 0.2826 0.2856 0.2902 0.2859 0.277
148 Reactions and Ratios
Genomic and Precision Medicine
0.2891 0.2854 0.2897 0.284 0.2858 Epigenetic Control of Gene
Expression
0.3367 0.3655 0.3616 0.365 0.335 Take the Lead on Healthcare
Quality Improvement
0.3918 0.3994 0.391 0.4015 0.393 Surviving Disruptive
Technologies
0.4096 0.4243 0.4186 0.423 0.4131 Caries Management by Risk
Assessment CAMBRA
0.3761 0.4094 0.4214 0.3989 0.3604 Computational Neuroscience 0.3194 0.3186 0.3191 0.32 0.3044 Introduction to Data Science 0.2593 0.2675 0.266 0.2669 0.2646 Discrete Optimization 0.2894 0.3003 0.3028 0.2973 0.2938 Foundations of Virtual
Instruction
0.2665 0.2976 0.2856 0.2864 0.2477 Virology I How Viruses Work 0.2527 0.2495 0.2501 0.2432 0.2392 Edx Introduction to Computer
Science
0.2925 0.3037 0.3054 0.3043 0.2926 edx Big Data in Education 0.4669 0.4725 0.467 0.4734 0.4727 edx Cellular mechanisms of
brain function
0.1805 0.1757 0.1767 0.1772 0.1786 edx Introduction to
Programming with Java Part 1 Starting to Code with Java
0.1413 0.1415 0.1394 0.1384 0.1401
edx Programming Basics 0.3799 0.3738 0.3752 0.3714 0.3764 總平均 0.3176 0.3297 0.3292 0.3276 0.3081
平均 F1 值:與動詞 LDA 摘要方法呈現顯著差異
平均 F1 值:與名詞 LDA 摘要方法呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比動詞 LDA 摘要方法或名詞 LDA 摘要方法高
在 表 65 的 結 果 中 除 了 Big Data Science with the BD2K-LINCS Data Coordination and Integration Center 課程、Journalism Skills for Engaged Citizens 課 程、Virology I How Viruses Work 課程、edx Cellular mechanisms of brain function 課程和 edx Programming Basics 課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值皆比名詞 LDA 摘要方法或動詞 LDA 摘要方法低,其餘課程 的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值至少有一個摘要 方法比名詞 LDA 摘要方法和動詞 LDA 摘要方法高。
表 66. 與線上摘要器 SweSum 比較結果
課程名稱 TF TF-ISF TF-SF SweSum
149
The Hardware/Software Interface 0.3245 0.3244 0.3191 0.3189 Social Network Analysis 0.3578 0.3588 0.3535 0.358 Web Application Architectures 0.4539 0.4476 0.4532 0.4658 Audio Signal Processing for Music Applications 0.1897 0.201 0.1897 0.2035 Malicious Software and its Underground
Economy Two Sides to Every Story
0.3629 0.3585 0.3633 0.3637 Big Data Science with the BD2K-LINCS Data
Coordination and Integration Center
0.3398 0.3393 0.3389 0.3322 Experimental Methods in Systems Biology 0.3303 0.3354 0.3322 0.3358 Dynamical Modeling Methods for Systems
Biology
0.3513 0.3517 0.342 0.3408
The Brain and Space 0.3074 0.3026 0.3041 0.3096
Network Analysis in Systems Biology 0.3302 0.3283 0.3254 0.3203
automata 0.52 0.523 0.5178 0.515
Natural Language Processing 2013 0.2813 0.2742 0.2791 0.2538 Beginning Game Programming with C 0.3326 0.3325 0.3214 0.3069
Climate Change 0.3166 0.3203 0.3241 0.3214
Journalism Skills for Engaged Citizens 0.353 0.3515 0.3534 0.352 Algorithms Design and Analysis Part 1 0.3135 0.3123 0.315 0.2971 Algorithms Design and Analysis Part 2 0.3439 0.3421 0.3406 0.3319 Introduction to Chemistry Reactions and Ratios 0.2856 0.2902 0.2859 0.2849 Genomic and Precision Medicine 0.2854 0.2897 0.284 0.2855 Epigenetic Control of Gene Expression 0.3655 0.3616 0.365 0.3605 Take the Lead on Healthcare Quality
Improvement
0.3994 0.391 0.4015 0.3891 Surviving Disruptive Technologies 0.4243 0.4186 0.423 0.4287 Caries Management by Risk Assessment
CAMBRA
0.4094 0.4214 0.3989 0.4219 Computational Neuroscience 0.3186 0.3191 0.32 0.3337 Introduction to Data Science 0.2675 0.266 0.2669 0.2667 Discrete Optimization 0.3003 0.3028 0.2973 0.2993 Foundations of Virtual Instruction 0.2976 0.2856 0.2864 0.255 Virology I How Viruses Work 0.2495 0.2501 0.2432 0.2487 Edx Introduction to Computer Science 0.3037 0.3054 0.3043 0.292 edx Big Data in Education 0.4725 0.467 0.4734 0.4781 edx Cellular mechanisms of brain function 0.1757 0.1767 0.1772 0.176 edx Introduction to Programming with Java Part
1 Starting to Code with Java
0.1415 0.1394 0.1384 0.151 edx Programming Basics 0.3738 0.3752 0.3714 0.3674
總平均 0.3297 0.3292 0.3276 0.3262
平均 F1 值:與線上摘要器 SweSum 呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比線上摘要器 SweSum 高
在表 66 的結果中除了 Web Application Architectures 課程、Audio Signal
Processing for Music Applications 課程、Malicious Software and its Underground
Economy Two Sides to Every Story 課程、Experimental Methods in Systems Biology
課程、The Brain and Space 課程、Surviving Disruptive Technologies 課程、Caries
150
Management by Risk Assessment CAMBRA 課程和 Computational Neuroscience 課 程的 TF 方法、TF-ISF 方法和 TF-SF 方法平均 F1 值皆比線上摘要器 SweSum 低,
The Hardware/Software Interface 0.3126 0.3178 0.322 0.3078 0.3079 Social Network Analysis 0.3405 0.3432 0.3537 0.3364 0.3296 Web Application Architectures 0.4236 0.4424 0.4408 0.4286 0.4304 Audio Signal Processing for MusicApplications
0.1734 0.1934 0.1953 0.1833 0.1791 Malicious Software and its
Underground Economy Two Sides to Every Story
0.3572 0.3425 0.3478 0.3322 0.3409
Big Data Science with the BD2K-LINCS Data Coordination and Integration Center
0.3295 0.3306 0.3358 0.3257 0.331
Experimental Methods in Systems Biology
0.3241 0.3295 0.3363 0.3223 0.3332 Dynamical Modeling Methods for
Systems Biology
0.3341 0.3446 0.3507 0.3311 0.3347 The Brain and Space 0.2833 0.2913 0.2982 0.281 0.2978 Network Analysis in Systems
Biology
0.3122 0.324 0.3257 0.3181 0.3227
automata 0.5121 0.5271 0.5222 0.5222 0.5083
Natural Language Processing 2013 0.2717 0.2778 0.2772 0.2667 0.2808 Beginning Game Programming
with C
0.2788 0.3106 0.327 0.2915 0.2765 Climate Change 0.2987 0.3193 0.3179 0.3043 0.2851 Journalism Skills for Engaged
Citizens
0.3474 0.3422 0.342 0.3389 0.351 Algorithms Design and Analysis
Part 1
0.284 0.304 0.3112 0.2968 0.2876 Algorithms Design and Analysis
Part 2
0.3198 0.3403 0.3425 0.3356 0.3265 Introduction to Chemistry
Reactions and Ratios
0.2704 0.2799 0.2819 0.2618 0.2726 Genomic and Precision Medicine 0.2874 0.2876 0.2885 0.2801 0.2866 Epigenetic Control of Gene 0.338 0.3547 0.3616 0.3336 0.33
151 Expression
Take the Lead on Healthcare Quality Improvement
0.3901 0.388 0.3809 0.3764 0.3895 Surviving Disruptive Technologies 0.4158 0.3963 0.4105 0.3807 0.4041 Caries Management by Risk
Assessment CAMBRA
0.3657 0.3796 0.4079 0.3497 0.3429 Computational Neuroscience 0.3143 0.3142 0.3158 0.3049 0.3084 Introduction to Data Science 0.2598 0.2578 0.2621 0.2514 0.2571 Discrete Optimization 0.2928 0.2921 0.2984 0.279 0.2846 Foundations of Virtual Instruction 0.2769 0.2374 0.2644 0.2206 0.232 Virology I How Viruses Work 0.2429 0.2432 0.2442 0.2314 0.238 Edx Introduction to Computer
Science
0.2967 0.2982 0.3008 0.2908 0.2872 edx Big Data in Education 0.4708 0.4672 0.4648 0.4607 0.4682 edx Cellular mechanisms of brain
function
0.1777 0.1776 0.1749 0.1765 0.1788 edx Introduction to Programming
with Java Part 1 Starting to Code with Java
0.1292 0.1373 0.1379 0.1328 0.134
edx Programming Basics 0.3789 0.372 0.3755 0.372 0.3789 總平均 0.3155 0.3201 0.3247 0.3099 0.3126
平均 F1 值:與一般 LDA 摘要方法呈現顯著差異
平均 F1 值:與 PLSA 摘要方法呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比一般 LDA 摘要方法或 PLSA 摘要方法高
在表 67 的結果中除了 Malicious Software and its Underground Economy Two Sides to Every Story 課程、Natural Language Processing 2013 課程、Journalism Skills for Engaged Citizens 課程、Take the Lead on Healthcare Quality Improvement 課程、Surviving Disruptive Technologies 課程、edx Big Data in Education 課程、edx Cellular mechanisms of brain function 課程和 edx Programming Basics 課程的 TF 摘 要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值皆比 PLSA 摘要方法或一 般 LDA 摘要方法低,其餘課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要 方法平均 F1 值至少有一個摘要方法比 PLSA 摘要方法和一般 LDA 摘要方法高。
表 68. 與名詞 LDA 摘要方法和動詞 LDA 摘要方法比較結果
課程名稱 動詞
LDA
TF TF-ISF TF-SF 名詞
LDA The Hardware/Software Interface 0.3158 0.3178 0.322 0.3078 0.313 Social Network Analysis 0.3466 0.3432 0.3537 0.3364 0.3503 Web Application Architectures 0.4286 0.4424 0.4408 0.4286 0.4399 Audio Signal Processing for Music 0.1879 0.1934 0.1953 0.1833 0.1734
152 Applications
Malicious Software and its Underground Economy Two Sides to Every Story
0.3101 0.3425 0.3478 0.3322 0.31
Big Data Science with the BD2K-LINCS Data Coordination and Integration Center
0.3408 0.3306 0.3358 0.3257 0.3425
Experimental Methods in Systems Biology
0.3268 0.3295 0.3363 0.3223 0.3313 Dynamical Modeling Methods for
Systems Biology
0.337 0.3446 0.3507 0.3311 0.3476 The Brain and Space 0.2957 0.2913 0.2982 0.281 0.2933 Network Analysis in Systems
Biology
0.3174 0.324 0.3257 0.3181 0.3173
automata 0.5086 0.5271 0.5222 0.5222 0.5134
Natural Language Processing 2013 0.2711 0.2778 0.2772 0.2667 0 Beginning Game Programming
with C
0.2924 0.3106 0.327 0.2915 0.2988 Climate Change 0.312 0.3193 0.3179 0.3043 0.3028 Journalism Skills for Engaged
Citizens
0.3484 0.3422 0.342 0.3389 0.3554 Algorithms Design and Analysis
Part 1
0.2863 0.304 0.3112 0.2968 0.2879 Algorithms Design and Analysis
Part 2
0.32 0.3403 0.3425 0.3356 0.3166 Introduction to Chemistry
Reactions and Ratios
0.2826 0.2799 0.2819 0.2618 0.277 Genomic and Precision Medicine 0.2891 0.2876 0.2885 0.2801 0.2858 Epigenetic Control of Gene Surviving Disruptive Technologies 0.4096 0.3963 0.4105 0.3807 0.4131 Caries Management by Risk
Assessment CAMBRA
0.3761 0.3796 0.4079 0.3497 0.3604 Computational Neuroscience 0.3194 0.3142 0.3158 0.3049 0.3044 Introduction to Data Science 0.2593 0.2578 0.2621 0.2514 0.2646 Discrete Optimization 0.2894 0.2921 0.2984 0.279 0.2938 Foundations of Virtual Instruction 0.2665 0.2374 0.2644 0.2206 0.2477 Virology I How Viruses Work 0.2527 0.2432 0.2442 0.2314 0.2392 Edx Introduction to Computer
Science edx Introduction to Programming
with Java Part 1 Starting to Code with Java
0.1413 0.1373 0.1379 0.1328 0.1401
edx Programming Basics 0.3799 0.372 0.3755 0.372 0.3764 總平均 0.3176 0.3201 0.3247 0.3099 0.3081
平均 F1 值:與動詞 LDA 摘要方法呈現顯著差異
平均 F1 值:與名詞 LDA 摘要方法呈現顯著差異
153
紅色字體平均 F1 值:平均 F1 值未比動詞 LDA 摘要方法或名詞 LDA 摘要方法高
在 表 68 的 結 果 中 除 了 Big Data Science with the BD2K-LINCS Data Coordination and Integration Center 課程、Journalism Skills for Engaged Citizens 課 程、Genomic and Precision Medicine 課程、Introduction to Chemistry Reactions and Ratios 課程、Take the Lead on Healthcare Quality Improvement 課程、Surviving Disruptive Technologies 課程、Computational Neuroscience 課程、Introduction to Data Science 課程、Foundations of Virtual Instruction 課程、Virology I How Viruses Work 課程、edx Cellular mechanisms of brain function 課程、edx Introduction to Programming with Java Part 1 Starting to Code with Java 課程和 edx Programming Basics 課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值皆比 名詞 LDA 摘要方法或動詞 LDA 摘要方法低,其餘課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值至少有一個摘要方法比名詞 LDA 摘要方 法和動詞 LDA 摘要方法高。
表 69. 與線上摘要器 SweSum 比較結果
課程名稱 TF TF-ISF TF-SF SweSum
The Hardware/Software Interface 0.3178 0.322 0.3078 0.3189 Social Network Analysis 0.3432 0.3537 0.3364 0.358 Web Application Architectures 0.4424 0.4408 0.4286 0.4658 Audio Signal Processing for Music
Applications
0.1934 0.1953 0.1833 0.2035 Malicious Software and its Underground
Economy Two Sides to Every Story
0.3425 0.3478 0.3322 0.3637 Big Data Science with the BD2K-LINCS
Data Coordination and Integration Center
0.3306 0.3358 0.3257 0.3322 Experimental Methods in Systems Biology 0.3295 0.3363 0.3223 0.3358 Dynamical Modeling Methods for Systems
Biology
0.3446 0.3507 0.3311 0.3408
The Brain and Space 0.2913 0.2982 0.281 0.3096
Network Analysis in Systems Biology 0.324 0.3257 0.3181 0.3203
automata 0.5271 0.5222 0.5222 0.515
Natural Language Processing 2013 0.2778 0.2772 0.2667 0.2538 Beginning Game Programming with C 0.3106 0.327 0.2915 0.3069
Climate Change 0.3193 0.3179 0.3043 0.3214
Journalism Skills for Engaged Citizens 0.3422 0.342 0.3389 0.352 Algorithms Design and Analysis Part 1 0.304 0.3112 0.2968 0.2971
154
Algorithms Design and Analysis Part 2 0.3403 0.3425 0.3356 0.3319 Introduction to Chemistry Reactions and
Ratios
0.2799 0.2819 0.2618 0.2849 Genomic and Precision Medicine 0.2876 0.2885 0.2801 0.2855 Epigenetic Control of Gene Expression 0.3547 0.3616 0.3336 0.3605 Take the Lead on Healthcare Quality
Improvement
0.388 0.3809 0.3764 0.3891 Surviving Disruptive Technologies 0.3963 0.4105 0.3807 0.4287 Caries Management by Risk Assessment
CAMBRA
0.3796 0.4079 0.3497 0.4219 Computational Neuroscience 0.3142 0.3158 0.3049 0.3337 Introduction to Data Science 0.2578 0.2621 0.2514 0.2667 Discrete Optimization 0.2921 0.2984 0.279 0.2993 Foundations of Virtual Instruction 0.2374 0.2644 0.2206 0.255 Virology I How Viruses Work 0.2432 0.2442 0.2314 0.2487 Edx Introduction to Computer Science 0.2982 0.3008 0.2908 0.292 edx Big Data in Education 0.4672 0.4648 0.4607 0.4781 edx Cellular mechanisms of brain function 0.1776 0.1749 0.1765 0.176 edx Introduction to Programming with Java
Part 1 Starting to Code with Java
0.1373 0.1379 0.1328 0.151 edx Programming Basics 0.372 0.3755 0.372 0.3674
總平均 0.3201 0.3247 0.3099 0.3262
平均 F1 值:與線上摘要器 SweSum 呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比線上摘要器 SweSum 高
在表 69 的結果中除了 Social Network Analysis 課程、Web Application Architectures 課程、Audio Signal Processing for Music Applications 課程、Malicious Software and its Underground Economy Two Sides to Every Story 課程、The Brain and Space 課程、Climate Change 課程、Surviving Disruptive Technologies 課程、
Caries Management by Risk Assessment CAMBRA 課 程 、 Computational Neuroscience 課 程 、 edx Big Data in Education 課 程 和 edx Introduction to Programming with Java Part 1 Starting to Code with Java 課程的 TF 方法、TF-ISF 方法和 TF-SF 方法平均 F1 值皆比線上摘要器 SweSum 低,其餘課程的 TF 摘要 方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值至少有一個摘要方法比線上 摘要器 SweSum 高。
未刪章節使用提示詞 TF 權重正規化加權提示詞時各摘要方法與 PLSA 摘要
方法和一般摘要方法比較結果如表 70 所示,與動詞 LDA 摘要方法和名詞 LDA
摘要方法比較結果如表 71 所示,與線上摘要器 SweSum 比較結果如表 72 所示。
155
表 70. 與 PLSA 摘要方法和一般摘要方法比較結果
課程名稱 一般
LDA
TF TF-ISF TF-SF PLSA The Hardware/Software Interface 0.3126 0.324 0.3241 0.3188 0.3079 Social Network Analysis 0.3405 0.3486 0.3492 0.3447 0.3296 Web Application Architectures 0.4236 0.4167 0.4139 0.4112 0.4304 Audio Signal Processing for Music
Applications
0.1734 0.173 0.1828 0.1721 0.1791 Malicious Software and its Underground
Economy Two Sides to Every Story
0.3572 0.3243 0.3243 0.3242 0.3409 Big Data Science with the
BD2K-LINCS Data Coordination and Integration Center
0.3295 0.3454 0.3463 0.3445 0.331
Experimental Methods in Systems Biology
0.3241 0.0876 0.0892 0.0858 0.3332 Dynamical Modeling Methods for
Systems Biology
0.3341 0.3784 0.3794 0.3677 0.3347 The Brain and Space 0.2833 0.3021 0.301 0.2975 0.2978 Network Analysis in Systems Biology 0.3122 0.3357 0.3353 0.3307 0.3227
automata 0.5121 0.5044 0.5065 0.5 0.5083
Natural Language Processing 2013 0.2717 0.2815 0.2741 0.2758 0.2808 Beginning Game Programming with C 0.2788 0.3106 0.3135 0.3015 0.2765 Climate Change 0.2987 0.3241 0.3243 0.3279 0.2851 Journalism Skills for Engaged Citizens 0.3474 0.3541 0.3518 0.3548 0.351 Algorithms Design and Analysis Part 1 0.284 0.3125 0.3123 0.3125 0.2876 Algorithms Design and Analysis Part 2 0.3198 0.3441 0.3421 0.3412 0.3265 Introduction to Chemistry Reactions and
Ratios
0.2704 0.2234 0.2215 0.2186 0.2726 Genomic and Precision Medicine 0.2874 0.2858 0.2893 0.2846 0.2866 Epigenetic Control of Gene Expression 0.338 0.3661 0.3606 0.3624 0.33 Take the Lead on Healthcare Quality
Improvement
0.3901 0.3978 0.3913 0.3983 0.3895 Surviving Disruptive Technologies 0.4158 0.4236 0.419 0.4136 0.4041 Caries Management by Risk Assessment
CAMBRA
0.3657 0.4086 0.4195 0.3933 0.3429 Computational Neuroscience 0.3143 0.318 0.3182 0.3182 0.3084 Introduction to Data Science 0.2598 0.2662 0.2662 0.2662 0.2571 Discrete Optimization 0.2928 0.3005 0.3027 0.2963 0.2846 Foundations of Virtual Instruction 0.2769 0.2871 0.2856 0.2883 0.232 Virology I How Viruses Work 0.2429 0.2487 0.2502 0.2412 0.238 Edx Introduction to Computer Science 0.2967 0.3035 0.3054 0.3016 0.2872 edx Big Data in Education 0.4708 0.4725 0.4669 0.473 0.4682 edx Cellular mechanisms of brain
function
0.1777 0.1756 0.1762 0.1767 0.1788 edx Introduction to Programming with
Java Part 1 Starting to Code with Java
0.1292 0.1414 0.1418 0.1388 0.134 edx Programming Basics 0.3789 0.3739 0.3751 0.3715 0.3789
總平均 0.3155 0.317 0.317 0.3137 0.3126
平均 F1 值:與一般 LDA 摘要方法呈現顯著差異
156 平均 F1 值:與 PLSA 摘要方法呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比一般 LDA 摘要方法或 PLSA 摘要方法高
在表 70 的文字搞摘要擷取結果中除了 Web Application Architectures 課程、
Malicious Software and its Underground Economy Two Sides to Every Story 課程、
Experimental Methods in Systems Biology 課程、automata 課程、Introduction to Chemistry Reactions and Ratios 課程、edx Cellular mechanisms of brain function 課 程和 edx Programming Basics 課程的 TF 方法、TF-ISF 方法和 TF-SF 方法平均 F1 值皆比一般 LDA 摘要方法或 PLSA 摘要方法低,其餘課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值至少有一個摘要方法比一般 LDA 摘要方 法和 PLSA 摘要方法高。
表 71. 與動詞 LDA 摘要方法和名詞 LDA 摘要方法比較結果
課程名稱 動詞
LDA
TF TF-ISF TF-SF 名詞 LDA The Hardware/Software Interface 0.3158 0.324 0.3241 0.3188 0.313 Social Network Analysis 0.3466 0.3486 0.3492 0.3447 0.3503 Web Application Architectures 0.4286 0.4167 0.4139 0.4112 0.4399 Audio Signal Processing for Music
Applications
0.1879 0.173 0.1828 0.1721 0.1734 Malicious Software and its Underground
Economy Two Sides to Every Story
0.3101 0.3243 0.3243 0.3242 0.31 Big Data Science with the
BD2K-LINCS Data Coordination and Integration Center
0.3408 0.3454 0.3463 0.3445 0.3425
Experimental Methods in Systems Biology
0.3268 0.0876 0.0892 0.0858 0.3313 Dynamical Modeling Methods for
Systems Biology
0.337 0.3784 0.3794 0.3677 0.3476 The Brain and Space 0.2957 0.3021 0.301 0.2975 0.2933 Network Analysis in Systems Biology 0.3174 0.3357 0.3353 0.3307 0.3173
automata 0.5086 0.5044 0.5065 0.5 0.5134
Natural Language Processing 2013 0.2711 0.2815 0.2741 0.2758 0 Beginning Game Programming with C 0.2924 0.3106 0.3135 0.3015 0.2988 Climate Change 0.312 0.3241 0.3243 0.3279 0.3028 Journalism Skills for Engaged Citizens 0.3484 0.3541 0.3518 0.3548 0.3554 Algorithms Design and Analysis Part 1 0.2863 0.3125 0.3123 0.3125 0.2879 Algorithms Design and Analysis Part 2 0.32 0.3441 0.3421 0.3412 0.3166 Introduction to Chemistry Reactions and
Ratios
0.2826 0.2234 0.2215 0.2186 0.277
157
Genomic and Precision Medicine 0.2891 0.2858 0.2893 0.2846 0.2858 Epigenetic Control of Gene Expression 0.3367 0.3661 0.3606 0.3624 0.335 Take the Lead on Healthcare Quality
Improvement
0.3918 0.3978 0.3913 0.3983 0.393 Surviving Disruptive Technologies 0.4096 0.4236 0.419 0.4136 0.4131 Caries Management by Risk Assessment
CAMBRA
0.3761 0.4086 0.4195 0.3933 0.3604 Computational Neuroscience 0.3194 0.318 0.3182 0.3182 0.3044 Introduction to Data Science 0.2593 0.2662 0.2662 0.2662 0.2646 Discrete Optimization 0.2894 0.3005 0.3027 0.2963 0.2938 Foundations of Virtual Instruction 0.2665 0.2871 0.2856 0.2883 0.2477 Virology I How Viruses Work 0.2527 0.2487 0.2502 0.2412 0.2392 Edx Introduction to Computer Science 0.2925 0.3035 0.3054 0.3016 0.2926 edx Big Data in Education 0.4669 0.4725 0.4669 0.473 0.4727 edx Cellular mechanisms of brain
function
0.1805 0.1756 0.1762 0.1767 0.1786 edx Introduction to Programming with
Java Part 1 Starting to Code with Java
0.1413 0.1414 0.1418 0.1388 0.1401 edx Programming Basics 0.3799 0.3739 0.3751 0.3715 0.3764
總平均 0.3176 0.317 0.317 0.3137 0.3081
平均 F1 值:與動詞 LDA 摘要方法呈現顯著差異
平均 F1 值:與名詞 LDA 摘要方法呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比動詞 LDA 摘要方法或名詞 LDA 摘要方法高
在表 71 的結果中除了 Social Network Analysis 課程、Web Application Architectures 課程、Audio Signal Processing for Music Applications 課程、Malicious Software and its Underground Economy Two Sides to Every Story 課 程 、 Experimental Methods in Systems Biology 課程、automata 課程、Journalism Skills for Engaged Citizens 課程、Introduction to Chemistry Reactions and Ratios 課程、
Computational Neuroscience 課程、Virology I How Viruses Work 課程、edx Cellular mechanisms of brain function 課程和 edx Programming Basics 課程的 TF 方法、
TF-ISF 方法和 TF-SF 方法平均 F1 值皆比動詞 LDA 摘要方法或名詞 LDA 摘要方 法低,其餘課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值 至少有一個摘要方法比動詞 LDA 摘要方法和名詞 LDA 摘要方法高。
表 72. 與線上摘要器 SweSum 比較結果
課程名稱 TF TF-ISF TF-SF SweSum
The Hardware/Software Interface 0.324 0.3241 0.3188 0.3189
158
Social Network Analysis 0.3486 0.3492 0.3447 0.358 Web Application Architectures 0.4167 0.4139 0.4112 0.4658 Audio Signal Processing for Music Applications 0.173 0.1828 0.1721 0.2035 Malicious Software and its Underground
Economy Two Sides to Every Story
0.3243 0.3243 0.3242 0.3637 Big Data Science with the BD2K-LINCS Data
Coordination and Integration Center
0.3454 0.3463 0.3445 0.3322 Experimental Methods in Systems Biology 0.0876 0.0892 0.0858 0.3358 Dynamical Modeling Methods for Systems
Biology
0.3784 0.3794 0.3677 0.3408
The Brain and Space 0.3021 0.301 0.2975 0.3096
Network Analysis in Systems Biology 0.3357 0.3353 0.3307 0.3203
automata 0.5044 0.5065 0.5 0.515
Natural Language Processing 2013 0.2815 0.2741 0.2758 0.2538 Beginning Game Programming with C 0.3106 0.3135 0.3015 0.3069
Climate Change 0.3241 0.3243 0.3279 0.3214
Journalism Skills for Engaged Citizens 0.3541 0.3518 0.3548 0.352 Algorithms Design and Analysis Part 1 0.3125 0.3123 0.3125 0.2971 Algorithms Design and Analysis Part 2 0.3441 0.3421 0.3412 0.3319 Introduction to Chemistry Reactions and Ratios 0.2234 0.2215 0.2186 0.2849 Genomic and Precision Medicine 0.2858 0.2893 0.2846 0.2855 Epigenetic Control of Gene Expression 0.3661 0.3606 0.3624 0.3605 Take the Lead on Healthcare Quality
Improvement
0.3978 0.3913 0.3983 0.3891 Surviving Disruptive Technologies 0.4236 0.419 0.4136 0.4287 Caries Management by Risk Assessment
CAMBRA
0.4086 0.4195 0.3933 0.4219 Computational Neuroscience 0.318 0.3182 0.3182 0.3337 Introduction to Data Science 0.2662 0.2662 0.2662 0.2667 Discrete Optimization 0.3005 0.3027 0.2963 0.2993 Foundations of Virtual Instruction 0.2871 0.2856 0.2883 0.255 Virology I How Viruses Work 0.2487 0.2502 0.2412 0.2487 Edx Introduction to Computer Science 0.3035 0.3054 0.3016 0.292 edx Big Data in Education 0.4725 0.4669 0.473 0.4781 edx Cellular mechanisms of brain function 0.1756 0.1762 0.1767 0.176 edx Introduction to Programming with Java Part
1 Starting to Code with Java
0.1414 0.1418 0.1388 0.151 edx Programming Basics 0.3739 0.3751 0.3715 0.3674
總平均 0.317 0.317 0.3137 0.3262
平均 F1 值:與線上摘要器 SweSum 呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比線上摘要器 SweSum 高
在表 72 的結果中除了 Social Network Analysis 課程、Web Application
Architectures 課程、Audio Signal Processing for Music Applications 課程、Malicious
Software and its Underground Economy Two Sides to Every Story 課 程 、
Experimental Methods in Systems Biology 課程、The Brain and Space 課程、automata
159
課程、Journalism Skills for Engaged Citizens 課程、Introduction to Chemistry Reactions and Ratios 課 程 、 Surviving Disruptive Technologies 課 程 、 Caries Management by Risk Assessment CAMBRA 課程、Introduction to Data Science 課程、
edx Big Data in Education 課程、edx Cellular mechanisms of brain function 課程、
edx Introduction to Programming with Java Part 1 Starting to Code with Java 課程和 edx Programming Basics 課程的 TF 方法、TF-ISF 方法和 TF-SF 方法平均 F1 值皆 比線上摘要器 SweSum 低,其餘課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值至少摘要有一個方法比線上摘要器 SweSum 高。
刪章節未加入提示詞時各摘要方法與 PLSA 摘要方法和一般摘要方法比較 結果如表 73 所示,與動詞 LDA 摘要方法和名詞 LDA 摘要方法比較結果如表 74 所示,與線上摘要器 SweSum 比較結果如表 75 所示。
表 73. 與 PLSA 摘要方法和一般 LDA 摘要方法比較結果
課程名稱 一般
LDA
TF TF-ISF TF-SF PLSA The Hardware/Software Interface 0.3123 0.3261 0.3271 0.3219 0.3151 Social Network Analysis 0.3543 0.3612 0.3597 0.3545 0.3322 Web Application Architectures 0.4432 0.4546 0.4549 0.4508 0.4318 Audio Signal Processing for
Music Applications
0.1734 0.1897 0.201 0.201 0.1791 Malicious Software and its
Underground Economy Two Sides to Every Story
0.3539 0.3582 0.3559 0.3617 0.3361
Big Data Science with the BD2K-LINCS Data Coordination and Integration Center
0.2981 0.3055 0.3081 0.2994 0.2849
Experimental Methods in Systems Biology
0.3241 0.3303 0.3354 0.3322 0.3332 Dynamical Modeling Methods
for Systems Biology
0.3412 0.3536 0.3548 0.3455 0.3309 The Brain and Space 0.2833 0.3074 0.3026 0.3041 0.2978 Network Analysis in Systems
Biology
0.3145 0.3193 0.3232 0.3165 0.3151
automata 0.5121 0.52 0.523 0.5178 0.5083
Natural Language Processing 2013
0.2678 0.2802 0.2744 0.2773 0.2664 Beginning Game Programming
with C
0.2788 0.3326 0.3325 0.3214 0.2765
160
Climate Change 0.2987 0.3166 0.3203 0.3241 0.2851 Journalism Skills for Engaged
Citizens
0.3491 0.3473 0.346 0.3454 0.3264 Algorithms Design and Analysis
Part 1
0.284 0.3135 0.3123 0.315 0.2876 Algorithms Design and Analysis
Part 2
0.3198 0.3439 0.3421 0.3406 0.3265 Introduction to Chemistry
Reactions and Ratios
0.2841 0.2906 0.2928 0.2872 0.275 Genomic and Precision Medicine 0.2793 0.2778 0.2797 0.2736 0.2672 Epigenetic Control of Gene Surviving Disruptive
Technologies
0.4028 0.4145 0.4089 0.4145 0.3892 Caries Management by Risk
Assessment CAMBRA
0.3657 0.4094 0.4214 0.3989 0.3429 Computational Neuroscience 0.3143 0.3186 0.3191 0.32 0.3084 Introduction to Data Science 0.2511 0.2606 0.2633 0.2634 0.2563 Discrete Optimization 0.2928 0.3003 0.3028 0.2973 0.2846 Foundations of Virtual Instruction 0.2769 0.2976 0.2856 0.2864 0.232 Virology I How Viruses Work 0.2343 0.2436 0.2444 0.2388 0.2282 Edx Introduction to Computer
Science
0.2886 0.2985 0.2981 0.2991 0.276 edx Big Data in Education 0.4632 0.4636 0.4553 0.4638 0.4484 edx Cellular mechanisms of brain
function
0.1777 0.1757 0.1767 0.1772 0.1788 edx Introduction to Programming
with Java Part 1 Starting to Code with Java
0.1262 0.1281 0.1267 0.122 0.1324
edx Programming Basics 0.3772 0.3705 0.3715 0.3674 0.3794 總平均 0.3145 0.3268 0.3266 0.3246 0.3078
平均 F1 值:與一般 LDA 摘要方法呈現顯著差異
平均 F1 值:與 PLSA 摘要方法呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比一般 LDA 摘要方法或 PLSA 摘要方法高
在表 73 的結果中除了 Journalism Skills for Engaged Citizens 課程、edx
Cellular mechanisms of brain function 課程、edx Introduction to Programming with
Java Part 1 Starting to Code with Java 課程和 edx Programming Basics 課程的 TF 摘
要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值皆比 PLSA 摘要方法或一
般 LDA 摘要方法低,其餘課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要
方法平均 F1 值至少有一個摘要方法比 PLSA 摘要方法和一般 LDA 摘要方法高。
161 The Hardware/Software Interface 0.316 0.3261 0.3271 0.3219 0.3171 Social Network Analysis 0.3465 0.3612 0.3597 0.3545 0.3461 Web Application Architectures 0.4334 0.4546 0.4549 0.4508 0.4337 Audio Signal Processing for
Music Applications
0.1879 0.1897 0.201 0.201 0.1734 Malicious Software and its
Underground Economy Two Sides to Every Story
0.3488 0.3582 0.3559 0.3617 0.3508
Big Data Science with the BD2K-LINCS Data Coordination and Integration Center
0.2933 0.3055 0.3081 0.2994 0.2988
Experimental Methods in Systems Biology
0.3268 0.3303 0.3354 0.3322 0.3314 Dynamical Modeling Methods
for Systems Biology
0.3347 0.3536 0.3548 0.3455 0.3479 The Brain and Space 0.2957 0.3074 0.3026 0.3041 0.2933 Network Analysis in Systems
Biology
0.312 0.3193 0.3232 0.3165 0.3174
automata 0.5086 0.52 0.523 0.5178 0.5134
Natural Language Processing 2013
0.2694 0.2802 0.2744 0.2773 0 Beginning Game Programming
with C
0.2924 0.3326 0.3325 0.3214 0.2988 Climate Change 0.312 0.3166 0.3203 0.3241 0.3029 Journalism Skills for Engaged
Citizens
0.3411 0.3473 0.346 0.3454 0.3332 Algorithms Design and Analysis
Part 1
0.2863 0.3135 0.3123 0.315 0.2879 Algorithms Design and Analysis
Part 2
0.3201 0.3439 0.3421 0.3406 0.3166 Introduction to Chemistry
Reactions and Ratios
0.2773 0.2906 0.2928 0.2872 0.2783 Genomic and Precision Medicine 0.2588 0.2778 0.2797 0.2736 0.2637 Epigenetic Control of Gene Surviving Disruptive
Technologies
0.4028 0.4145 0.4089 0.4145 0.407 Caries Management by Risk
Assessment CAMBRA
0.3761 0.4094 0.4214 0.3989 0.3604 Computational Neuroscience 0.3194 0.3186 0.3191 0.32 0.3044 Introduction to Data Science 0.2506 0.2606 0.2633 0.2634 0.2504 Discrete Optimization 0.2895 0.3003 0.3028 0.2973 0.2938 Foundations of Virtual Instruction 0.2665 0.2976 0.2856 0.2864 0.2478 Virology I How Viruses Work 0.2343 0.2436 0.2444 0.2388 0.2208 Edx Introduction to Computer
Science
0.2836 0.2985 0.2981 0.2991 0.287
162
edx Big Data in Education 0.453 0.4636 0.4553 0.4638 0.4616 edx Cellular mechanisms of brain
function
0.1805 0.1757 0.1767 0.1772 0.1786 edx Introduction to Programming
with Java Part 1 Starting to Code with Java
0.1223 0.1281 0.1267 0.122 0.1321
edx Programming Basics 0.3702 0.3705 0.3715 0.3674 0.3777 總平均 0.3134 0.3268 0.3266 0.3246 0.3048
平均 F1 值:與動詞 LDA 摘要方法呈現顯著差異
平均 F1 值:與名詞 LDA 摘要方法呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比動詞 LDA 摘要方法或名詞 LDA 摘要方法高
在表 74 的文字搞摘要擷取結果中除了 edx Cellular mechanisms of brain function 課程、edx Introduction to Programming with Java Part 1 Starting to Code with Java 課程和 edx Programming Basics 課程的 TF 摘要方法、TF-ISF 摘要方法 和 TF-SF 摘要方法平均 F1 值皆比動詞 LDA 摘要方法或名詞 LDA 摘要方法低,
其餘課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值至少有 一個摘要方法比名詞 LDA 摘要方法和動詞 LDA 摘要方法高。
表 75. 與線上摘要器 SweSum 比較結果
課程名稱 TF TF-ISF TF-SF SweSum
The Hardware/Software Interface 0.3261 0.3271 0.3219 0.321 Social Network Analysis 0.3612 0.3597 0.3545 0.3598 Web Application Architectures 0.4546 0.4549 0.4508 0.4658 Audio Signal Processing for Music Applications 0.1897 0.201 0.201 0.2035 Malicious Software and its Underground
Economy Two Sides to Every Story
0.3582 0.3559 0.3617 0.3635 Big Data Science with the BD2K-LINCS Data
Coordination and Integration Center
0.3055 0.3081 0.2994 0.2976 Experimental Methods in Systems Biology 0.3303 0.3354 0.3322 0.3358 Dynamical Modeling Methods for Systems
Biology
0.3536 0.3548 0.3455 0.3425 The Brain and Space 0.3074 0.3026 0.3041 0.3096 Network Analysis in Systems Biology 0.3193 0.3232 0.3165 0.3132
automata 0.52 0.523 0.5178 0.515
Natural Language Processing 2013 0.2802 0.2744 0.2773 0.2541 Beginning Game Programming with C 0.3326 0.3325 0.3214 0.3069
Climate Change 0.3166 0.3203 0.3241 0.3214
Journalism Skills for Engaged Citizens 0.3473 0.346 0.3454 0.3472 Algorithms Design and Analysis Part 1 0.3135 0.3123 0.315 0.2971 Algorithms Design and Analysis Part 2 0.3439 0.3421 0.3406 0.3319 Introduction to Chemistry Reactions and Ratios 0.2906 0.2928 0.2872 0.2893
163
Genomic and Precision Medicine 0.2778 0.2797 0.2736 0.2741 Epigenetic Control of Gene Expression 0.3655 0.3616 0.365 0.3605 Take the Lead on Healthcare Quality
Improvement
0.4084 0.3968 0.4076 0.3959 Surviving Disruptive Technologies 0.4145 0.4089 0.4145 0.4195 Caries Management by Risk Assessment
CAMBRA
0.4094 0.4214 0.3989 0.4219 Computational Neuroscience 0.3186 0.3191 0.32 0.3337 Introduction to Data Science 0.2606 0.2633 0.2634 0.2597 Discrete Optimization 0.3003 0.3028 0.2973 0.2993 Foundations of Virtual Instruction 0.2976 0.2856 0.2864 0.255 Virology I How Viruses Work 0.2436 0.2444 0.2388 0.2454 Edx Introduction to Computer Science 0.2985 0.2981 0.2991 0.2832 edx Big Data in Education 0.4636 0.4553 0.4638 0.2605 edx Cellular mechanisms of brain function 0.1757 0.1767 0.1772 0.176 edx Introduction to Programming with Java Part
1 Starting to Code with Java
0.1281 0.1267 0.122 0.122 edx Programming Basics 0.3705 0.3715 0.3674 0.1559
總平均 0.3268 0.3266 0.3246 0.3102
平均 F1 值:與線上摘要器 SweSum 呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比線上摘要器 SweSum 高
在表 75 的結果中除了 Web Application Architectures 課程、Audio Signal Processing for Music Applications 課程、Malicious Software and its Underground Economy Two Sides to Every Story 課程、Experimental Methods in Systems Biology 課程、The Brain and Space 課程、Surviving Disruptive Technologies 課程、Caries Management by Risk Assessment CAMBRA 課程、Computational Neuroscience 課 程和 Virology I How Viruses Work 課程的 TF 方法、TF-ISF 方法和 TF-SF 方法平 均 F1 值皆比線上摘要器 SweSum 低,其餘課程的 TF 摘要方法、TF-ISF 摘要方 法和 TF-SF 摘要方法平均 F1 值至少有一個摘要方法比線上摘要器 SweSum 高。
刪章節使用提示詞 TF 權重正規化加權提示詞時各摘要方法與 PLSA 摘要方 法和一般摘要方法比較結果如表 76 所示,與動詞 LDA 摘要方法和名詞 LDA 摘 要方法比較結果如表 77 所示,與線上摘要器 SweSum 比較結果如表 78 所示。
表 76. 與 PLSA 摘要方法和一般 LDA 摘要方法比較結果
課程名稱 一般
LDA
TF TF-ISF TF-SF PLSA
164
The Hardware/Software Interface 0.3123 0.3185 0.3232 0.31 0.3151 Social Network Analysis 0.3543 0.3466 0.3521 0.3368 0.3322 Web Application Architectures 0.4432 0.4417 0.4411 0.4286 0.4318 Audio Signal Processing for
Music Applications
0.1734 0.1934 0.1953 0.1833 0.1791 Malicious Software and its
Underground Economy Two Sides to Every Story
0.3539 0.3433 0.3492 0.3292 0.3361
Big Data Science with the BD2K-LINCS Data Coordination and Integration Center
0.2981 0.2963 0.3075 0.2851 0.2849
Experimental Methods in Systems Biology
0.3241 0.3295 0.3363 0.3223 0.3332 Dynamical Modeling Methods
for Systems Biology
0.3412 0.3446 0.3507 0.3311 0.3309 The Brain and Space 0.2833 0.2913 0.2982 0.281 0.2978 Network Analysis in Systems
Biology
0.3145 0.3171 0.3199 0.3071 0.3151
automata 0.5121 0.5271 0.5222 0.5222 0.5083
Natural Language Processing 2013
0.2678 0.2772 0.277 0.268 0.2664 Beginning Game Programming
with C
0.2788 0.3106 0.327 0.2915 0.2765 Climate Change 0.2987 0.3193 0.3179 0.3043 0.2851 Journalism Skills for Engaged
Citizens
0.3491 0.3335 0.3356 0.3274 0.3264 Algorithms Design and Analysis
Part 1
0.284 0.304 0.3112 0.2968 0.2876 Algorithms Design and Analysis
Part 2
0.3198 0.3403 0.3425 0.3356 0.3265 Introduction to Chemistry
Reactions and Ratios
0.2841 0.2848 0.2938 0.2646 0.275 Genomic and Precision Medicine 0.2793 0.2744 0.2789 0.268 0.2672 Epigenetic Control of Gene Surviving Disruptive
Technologies
0.4028 0.3873 0.4009 0.3708 0.3892 Caries Management by Risk
Assessment CAMBRA
0.3657 0.3796 0.4079 0.3497 0.3429 Computational Neuroscience 0.3143 0.3142 0.3158 0.3049 0.3084 Introduction to Data Science 0.2511 0.2578 0.2621 0.2514 0.2563 Discrete Optimization 0.2928 0.2921 0.2984 0.279 0.2846 Foundations of Virtual Instruction 0.2769 0.2374 0.2644 0.2206 0.232 Virology I How Viruses Work 0.2343 0.2411 0.2386 0.2278 0.2282 Edx Introduction to Computer
Science
0.2886 0.289 0.2939 0.2818 0.276 edx Big Data in Education 0.4632 0.4554 0.4541 0.4466 0.4484 edx Cellular mechanisms of brain
function
0.1777 0.1776 0.1749 0.1765 0.1788 edx Introduction to Programming 0.1262 0.1245 0.1284 0.1216 0.1324
165 with Java Part 1 Starting to Code
with Java
edx Programming Basics 0.3772 0.3662 0.3712 0.3661 0.3794 總平均 0.3145 0.317 0.3225 0.3062 0.3078
平均 F1 值:與一般 LDA 摘要方法呈現顯著差異
平均 F1 值:與 PLSA 摘要方法呈現顯著差異
紅色字體平均 F1 值:平均 F1 值未比一般 LDA 摘要方法或 PLSA 摘要方法高
在表 76 的結果中除了 Malicious Software and its Underground Economy Two Sides to Every Story 課 程 、 Big Data Science with the BD2K-LINCS Data Coordination and Integration Center 課程、Network Analysis in Systems Biology 課 程、Natural Language Processing 2013 課程、Journalism Skills for Engaged Citizens 課程、Genomic and Precision Medicine 課程、Take the Lead on Healthcare Quality Improvement 課程、Surviving Disruptive Technologies 課程、Foundations of Virtual Instruction 課程、edx Cellular mechanisms of brain function 課程、edx Introduction to Programming with Java Part 1 Starting to Code with Java 課程和 edx Programming Basics 課程的 TF 摘要方法、TF-ISF 摘要方法和 TF-SF 摘要方法平均 F1 值皆比 PLSA 摘要方法或一般 LDA 摘要方法低,其餘課程的 TF 摘要方法、TF-ISF 摘 要方法和 TF-SF 摘要方法平均 F1 值至少有一個摘要方法比 PLSA 摘要方法和一 般 LDA 摘要方法高。
表 77. 與名詞 LDA 摘要方法和動詞 LDA 摘要方法比較結果
課程名稱 動詞
LDA
TF TF-ISF TF-SF 名詞
LDA The Hardware/Software Interface 0.316 0.3185 0.3232 0.31 0.3171 Social Network Analysis 0.3465 0.3466 0.3521 0.3368 0.3461 Web Application Architectures 0.4334 0.4417 0.4411 0.4286 0.4337 Audio Signal Processing for
Music Applications
0.1879 0.1934 0.1953 0.1833 0.1734 Malicious Software and its
Underground Economy Two Sides to Every Story
0.3488 0.3433 0.3492 0.3292 0.3508
Big Data Science with the BD2K-LINCS Data Coordination and Integration Center
0.2933 0.2963 0.3075 0.2851 0.2988
Experimental Methods in 0.3268 0.3295 0.3363 0.3223 0.3314
166 Systems Biology
Dynamical Modeling Methods for Systems Biology
0.3347 0.3446 0.3507 0.3311 0.3479 The Brain and Space 0.2957 0.2913 0.2982 0.281 0.2933 Network Analysis in Systems
Biology
0.312 0.3171 0.3199 0.3071 0.3174
automata 0.5086 0.5271 0.5222 0.5222 0.5134
Natural Language Processing 2013
0.2694 0.2772 0.277 0.268 0 Beginning Game Programming
with C
0.2924 0.3106 0.327 0.2915 0.2988 Climate Change 0.312 0.3193 0.3179 0.3043 0.3029 Journalism Skills for Engaged
Citizens
0.3411 0.3335 0.3356 0.3274 0.3332 Algorithms Design and Analysis
Part 1
0.2863 0.304 0.3112 0.2968 0.2879 Algorithms Design and Analysis
Part 2
0.3201 0.3403 0.3425 0.3356 0.3166 Introduction to Chemistry
Reactions and Ratios
0.2773 0.2848 0.2938 0.2646 0.2783 Genomic and Precision Medicine 0.2588 0.2744 0.2789 0.268 0.2637 Epigenetic Control of Gene Surviving Disruptive
Technologies
0.4028 0.3873 0.4009 0.3708 0.407 Caries Management by Risk
0.4028 0.3873 0.4009 0.3708 0.407 Caries Management by Risk