Pembimbing 1 : Diana Purwitasari, S.Kom, M.Sc. Pembimbing 2 : Umi Laily Yuhana, S.Kom, M.Sc.
Anotasi = rangkuman / catatan / penandaan Pada sebuah artikel anotasi = penanda topik-topik apa saja yang dibahas Banyak artikel e-learning belum dilengkapi anotasi Anotasi manual = menghabiskan banyak waktu
Membuat sistem yang dapat memberikan rekomendasi anotasi
Web e-learning tentang Software Engineering Artikel dibedakan dalam kategori-kategori Kategori berdasarkan SWEBOK Artikel baru rekomendasi anotasi berupa kategori lain yang ‘menurut sistem’ juga dibahas dalam artikel tersebut
Flowchart The first structured method for documenting process flow, the flow process chart, was introduced by Frank Gilberth to members of ASME in 1921 as the presentation Process Charts-First Steps in Finding the One Best Way. Gilbreths tools quickly found their way into industrial engineering curricula. In the early 1930s, an industrial engineer, Allan H. Mogensen began training business people in the use of some of the tools of industrial engineering at his Work Simplification Conferences in Lake Placid, New York. Flowcharts used to be a popular means for describing computer algorithms and are still used for this purpose. Modern techniques such as UML activity diagrams can be considered to be extensions of the flowchart. Back in the 1970s the popularity of flowcharts as an own method decreased when interactive computer terminals and third-generation programming languages became the common tools of the trade, since algorithms can be expressed much more concisely and readably as source code in such a language. Often, pseudo-code is used, which uses the common idioms of such languages without strictly adhering to the details of a particular one.
A Data Flow Diagram provides no information about the timing or ordering of processes, or about whether processes will operate in sequence or in parallel. It is therefore quite different from a flowchart, which shows the flow of control through an algorithm, allowing a reader to determine what operations will be performed, in what order, and under what circumstances, but not what kinds of data will be input to and output from the system, nor where the data will come from and go to, nor where the data will be stored (all of which are shown on a DFD).
Kategori Root : Software Engineering.
Ekstraksi Fitur Term ((Pendataan Pendataan Kata Kata))
Oracle Text Basic Lexer
FLOW, FLOWCHART, METHOD, dan TOOLS
Pembobotan tf-idf Pembobotan tf-idf (term frequency– inverse document frequency) Pengukuran statistik pentingnya sebuah kata ke dokumen Menyeleksi kata dari hasil pendataan kata 30% bobot tertinggi
Pembobotan tf-idf Term
Jumlah Bobot
Flow
4.0
Flowchart
4.0
Method
4.0
Tools
4.0
Kata Penting : FLOW, FLOWCHART, METHOD, dan TOOLS
Ontology
Contoh hubungan kategori
Kategori Relasi Kategori Relasi Software Engineering Body of Knowledge Software Design Strategies and Methods Software Design Quality Analysis and Evaluation Software Design Notations Software Design Fundamentals Software Structure and Architecture
Key Issues in Software Design Software Requirements Software Maintenance Software Quality Software Engineering Tools and Methods Software Engineering Management Software Testing Software Configuration Management Software Engineering Process Software Construction Knowledge Areas of The Related Disciplines
Latent Semantic Analysis (LSA) LSA adalah sebuah teknik matematika/statistik Hubungan antara term/kata dan dokumen akan direpresentasikan ke dalam latent semantic space. Kata-kata penting dan artikel term document matrix LSA menggunakan metode SVD (Singular value decomposition) untuk memproses kata-kata yang ada dalam dokumen.
LSA
Didapatkan 3 nilai (scales, term vector, doc vector) Scales
Term vectors
Doc vectors
0 2,36
(-0,44, -0,60)
(-0,66, 0,22)
1 1,20
(-0,44, -0,51)
(-0,66, 0,30)
(-0,56, 0,43)
(-0,37, -0,93)
Kategori relasi akan dijadikan query untuk menghitung document scores
LSA No Artikel AR07
Kategori Relasi sebagai query
Query Vectors
Nilai kedekatan
Software Engineering Body of Knowledge
( 0.00, 0.00 )
0,0000
Software Design Strategies and Methods
( 0.00, 0.00 )
0,0000
Software Design Quality Analysis and Evaluation
( 0.00, 0.00 )
0,0000
Software Design Notations
( 0.00, 0.00 )
0,0000
Software Design Fundamentals
( 0.00, 0.00 )
0,0000
Software Structure and Architecture
( 0.00, 0.00 )
0,0000
Key Issues in Software Design
( 0.00, 0.00 )
0,0000
Software Requirements
( 0.00, 0.00 )
0,0000
Software Maintenance
( 0.00, 0.00 )
0,0000
Software Quality
( 0.00, 0.00 )
0,0000
Software Engineering Tools and Methods
( 0.56, -0.43 )
0,1177
Software Engineering Management
( 0.00, 0.00 )
0,0000
Software Testing
( 0.00, 0.00 )
0,0000
Software Configuration Management
( 0.00, 0.00 )
0,0000
Software Engineering Process
( 0.00, 0.00 )
0,0000
Software Construction
( 0.00, 0.00 )
0,0000
Knowledge Areas of The Related Disciplines
( 0.00, 0.00 )
0,0000
Wordnet Wordnet merupakan kamus (database leksikal) online untuk bahasa inggris Pada WordNet, terdapat sekumpulan taksonomi yang setiap node-nya diwakili oleh sebuah synset yang hanya direpresentasikan oleh satu sense
Wordnet
Wordnet
Untuk menghitung kemiripan formula Wu & Palmer.
“WordsMatching” yang dibuat oleh Thanh Ngoc Dao dan Troy Simpson menghitung kemiripan antar kalimat
Wordnet No Artikel AR07
Kategori Relasi
Nilai Kedekatan
Software Engineering Body of Knowledge
0,1634
Software Design Strategies and Methods
0,1769
Software Design Quality Analysis and Evaluation
0,1647
Software Design Notations
0,1625
Software Design Fundamentals
0,1493
Software Structure and Architecture
0,1568
Key Issues in Software Design
0,1653
Software Requirements
0,1242
Software Maintenance
0,1558
Software Quality
0,1192
Software Engineering Tools and Methods
0,1959
Software Engineering Management
0,1521
Software Testing
0,1463
Software Configuration Management
0,1418
Software Engineering Process
0,1761
Software Construction
0,1488
Knowledge Areas of The Related Disciplines
0,1650
No.
Judul Artikel
Rekomendasi
Presisi
1
Modelling Language
SWE02, SWE03, SWE04, SWE05, SWE06, SWE07, SWE08
86%
2
Software Design
SWE02, SWE03, SWE04, SWE05, SWE07
80%
3
Software design document
SWE02, SWE03, SWE04, SWE05, SWE06, SWE07
67%
4
Structured Systems Analysis and Design Method
SWE02, SWE03, SWE04, SWE05, SWE07
60%
5
Design Maker
SWE02, SWE03, SWE04, SWE05, SWE07
80%
6
Top-down and bottom-up design
SWE02
100%
7
Flowchart
SWE09
100%
8
Maintenance mode
SWE10, SWE11, SWE12, SWE13
75%
9
Software modernization
SWE01, SWE10, SWE11, SWE12, SWE13
60%
10
Software Maintenance
SWE10, SWE11, SWE12, SWE13, SWE14, SWE15, SWE16
71%
Metode Latent Semantic Analysis (LSA) dapat digunakan untuk memberikan rekomendasi anotasi dari suatu materi e-learning secara otomatis. Tetapi pada hasil rekomendasi ini dapat terjadi kondisi dimana tidak ada kategori yang direkomendasikan untuk jadi anotasi. Rekomendasi anotasi dapat juga didapatkan dengan bantuan Wordnet. Metode ini selalu menghasilkan rekomendasi, akan tetapi rekomendasi ini memiliki nilai presisi yang kurang baik yaitu 46.3%. Untuk mengatasi kondisi tidak adanya rekomendasi pada metode LSA, sistem akan memberikan rekomendasi dengan metode Wordnet apabila kondisi ini terjadi. Dan metode gabungan ini memberikan nilai presisi yang lebih baik yaitu 78.6%.
Kategori artikel bidang lain Dipadukan dengan sistem pencarian berbasis konteks Menggunakan metode lain dibandingkan nilai presisinya dengan metode LSA dan Wordnet pada tugas akhir ini.
