ebook img

universitas indonesia analisis opini konsumen berbasis fitur dalam bahasa indonesia PDF

86 Pages·2013·5.34 MB·Indonesian
by  
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview universitas indonesia analisis opini konsumen berbasis fitur dalam bahasa indonesia

UNIVERSITAS INDONESIA ANALISIS OPINI KONSUMEN BERBASIS FITUR DALAM BAHASA INDONESIA : STUDI KASUS PADA PRODUK GADGET E-COMMERCE KARYA AKHIR LISTIAN PRATOMO 1106042132 FAKULTAS ILMU KOMPUTER PROGRAM STUDI MAGISTER TEKNOLOGI INFORMASI JAKARTA JANUARI 2013 Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 UNIVERSITAS INDONESIA ANALISIS OPINI KONSUMEN BERBASIS FITUR DALAM BAHASA INDONESIA : STUDI KASUS PADA PRODUK GADGET E-COMMERCE KARYA AKHIR Diajukan sebagai salah satu syarat untuk memperoleh gelar Magister Teknologi Informasi LISTIAN PRATOMO 1106042132 FAKULTAS ILMU KOMPUTER PROGRAM STUDI MAGISTER TEKNOLOGI INFORMASI JAKARTA JANUARI 2013 Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 ii Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 iii Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 KATA PENGANTAR/UCAPAN TERIMA KASIH Puji syukur saya haturkan kepada Allah SWT yang telah memberikan berkat dan rahmat-Nya sehingga saya dapat menyelesaikan Karya Akhir ini. Saya menyadari, tanpa bantuan dari berbagai pihak, sangatlah sulit bagi saya untuk menyelesaikannya. Maka dari itu, saya mengucapkan terima kasih saya kepada : 1. Mama dan papa yang selalu memberikan kasih sayang, dukungan dan doa selama ini kepada penulis. Serta untuk adik-adikku Dimas dan Aji; 2. Ibu Yova Ruldeviyani, M.Kom, selaku dosen pembimbing yang selalu sabar menyediakan waktu membimbing penulis menyelesaikan Karya Akhir ini; 3. Bapak Edric Mandagi, beserta pihak PT. WEBARQ yang telah banyak membantu memberikan data penelitian Karya Akhir ini; 4. Dosen pengajar dan staf MTI UI yang telah memberikan banyak ilmu dan bantuan kepada saya; 5. Teman – teman MTI UI 2011SB, yang telah menghadirkan keluarga baru bagi saya saat di perkuliahan; dan 6. Seluruh pihak yang tidak dapat saya sebutkan satu per satu yang telah memberikan dukungan kepada saya dalam menyelesaikan Karya Akhir ini. Jakarta, 15 Januari 2013 Penulis iv Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 v Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 ABSTRAK Nama : Listian Pratomo Program Studi : Magister Teknologi Informasi Judul : Analisis Opini Konsumen Berbasis Fitur dalam Bahasa Indonesia : Studi Kasus pada Produk Gadget E-commerce Jumlah review mengalami peningkatan yang sangat pesat untuk setiap produk nya. Hal ini berakibat sulit nya bagi setiap pengguna untuk membaca semua review yang ada. Karya akhir ini menawarkan solusi menggunakan feature based opinion mining untuk mempermudah pengguna membaca review lebih mudah. Pada karya akhir ini terdapat 2 langkah yang akan dilakukan. Langkah pertama ialah melakukan ekstraksi feature menggunakan association rule dan pruning. Sedangkan langkah terakhir ialah menentukan orientasi dari setiap opini dengan menggunakan teknik klasifikasi. Beberapa algoritma klasifikasi seperti C45, Naïve Bayes dan Support Vector Machine cocok untuk mengatasi masalah ini. Dari hasil pengujian algoritma Support Vector Machine memiliki performa terbaik jika dibandingkan dengan algoritma lainnya. Kata kunci : Feature Extraction, Sentiment Analysis, C45, Naïve Bayes, SVM vi Universitas Indonesia Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 ABSTRACT Name : Listian Pratomo Program Study : Master of Information Technology Title : Analysis of Indonesian Feature Based Customer Opinion : Case Study in E-commerce Gadget Product The number of customer reviews for each product grows rapidly. This condition makes customer difficult to read all the review.This thesis propose feature based opinion mining to help customer reads review easily. Feature based opinion mining in this thesis consist of two steps. First step identify product features using association technique and pruning. The last step identify opinion sentence orientation using classification technique. Several classification algorithm, such as C45, Naive Bayes, and Support Vector Machines are good approaches to solve this problem. Support Vector Machine has the best performance compared to other algorithms. Keywords : Feature Extraction, Sentiment Analysis, C45, Naïve Bayes, SVM vii Universitas Indonesia Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 DAFTAR ISI HALAMAN PERNYATAAN ORISINALITAS ........................................................................II HALAMAN PENGESAHAN ...................................................................................................III KATA PENGANTAR/UCAPAN TERIMA KASIH ............................................................... IV HALAMAN PERNYATAAN PERSETUJUAN PUBLIKASI KARYA AKHIR UNTUK KEPENTINGAN AKADEMIS ................................................................................................. V ABSTRAK ............................................................................................................................... VI ABSTRACT ............................................................................................................................ VII DAFTAR ISI .......................................................................................................................... VIII DAFTAR TABEL ..................................................................................................................... X DAFTAR GAMBAR ................................................................................................................ XI BAB 1 PENDAHULUAN...................................................................................................... 1 1.1 LATAR BELAKANG....................................................................................................... 1 1.2 PERUMUSAN MASALAH ............................................................................................... 2 1.3 RUANG LINGKUP PENELITIAN ...................................................................................... 3 1.4 TUJUAN DAN MANFAAT PENELITIAN ............................................................................ 3 1.5 SISTEMATIKA PEMBAHASAN ........................................................................................ 3 BAB 2 LANDASAN TEORI ................................................................................................. 5 2.1 TEXT MINING .............................................................................................................. 5 2.1.1 POS Tagging .......................................................................................................... 5 2.1.2 StopWord ................................................................................................................ 7 2.1.3 Stemming ................................................................................................................ 7 2.1.4 Pembobotan ............................................................................................................ 7 2.2 FEATURE EXTRACTION ................................................................................................ 8 2.3 FP-GROWTH .............................................................................................................. 13 2.4 SENTIMENT ANALYSIS ............................................................................................... 14 2.5 C45 ........................................................................................................................... 17 2.6 NAÏVE BAYES ............................................................................................................ 20 2.7 SUPPORT VECTOR MACHINE ...................................................................................... 23 2.8 EVALUASI KLASIFIKASI ............................................................................................. 24 2.8.1 Confusion Matrix .................................................................................................. 24 2.8.2 Accuracy............................................................................................................... 25 2.8.3 Precision .............................................................................................................. 25 2.8.4 Recall ................................................................................................................... 25 2.8.5 F-Measure ............................................................................................................ 26 BAB 3 METODOLOGI PENELITIAN ............................................................................. 27 3.1 PERUMUSAN MASALAH ............................................................................................. 27 3.2 STUDI LITERATUR ...................................................................................................... 28 3.3 PENGUMPULAN DATA................................................................................................. 28 3.4 PENGUJIAN ................................................................................................................ 28 3.5 ANALISIS ................................................................................................................... 28 viii Universitas Indonesia Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013 3.6 HASIL PENELITIAN ..................................................................................................... 28 BAB 4 PEMROSESAN DATA ........................................................................................... 29 4.1 PROFIL PERUSAHAAN ................................................................................................. 29 4.2 TAHAPAN PEMROSESAN DATA ................................................................................... 29 4.2.1 Pengumpulan Data ............................................................................................... 30 4.2.2 POS Tagging ........................................................................................................ 30 4.2.3 Identifikasi Opini .................................................................................................. 32 4.2.4 Pencarian Frequent Feature ................................................................................. 35 4.2.5 Prunning............................................................................................................... 37 4.2.6 Pemberian Label Opini ......................................................................................... 38 4.2.7 Proses Stopword ................................................................................................... 38 4.2.8 Konversi ke Bentuk Vector Space Model ................................................................ 39 BAB 5 ANALISIS ............................................................................................................... 41 5.1 DATA ........................................................................................................................ 41 5.2 FEATURE EXTRACTION .............................................................................................. 41 5.2.1 Frequent Feature .................................................................................................. 42 5.2.2 Redundancy Prunning ........................................................................................... 44 5.3 KLASIFIKASI .............................................................................................................. 45 5.3.1 Pengujian menggunakan data asli ......................................................................... 46 5.3.2 Pengujian menggunakan data hasil overssampling ................................................ 49 5.3.3 Pengujian menggunakan data produk yang berbeda .............................................. 51 BAB 6 KESIMPULAN DAN SARAN ................................................................................ 54 6.1 KESIMPULAN ................................................................................................................ 54 6.2 SARAN ........................................................................................................................ 54 DAFTAR REFERENSI ........................................................................................................... 56 LAMPIRAN ............................................................................................................................. 59 1. PART OF SPEECH LABEL ................................................................................................. 59 2. STOPWORD .................................................................................................................... 60 3. AKTUAL FEATURE IPHONE 4S ........................................................................................ 64 4. AKTUAL FEATURE SAMSUNG GALAXY TAB 2.7.0 ........................................................ 65 5. FREQUENT FEATURE IPHONE 4S .................................................................................... 66 6. FREQUENT FEATURE SAMSUNG GALAXY TAB 2.7.0 .................................................... 68 7. FEATURE HASIL PRUNNING IPHONE 4S .......................................................................... 69 8. FEATURE HASIL PRUNNING SAMSUNG GALAXY TAB 2.7.0 .......................................... 71 9. CONTOH HASIL KLASIFIKASI .......................................................................................... 72 ix Universitas Indonesia Analisis opini ..., Yova Ruldeviyani, Fasilkom UI, 2013

Description:
Berdasarkan uraian singkat diatas, secara garis besar permasalahan yang harus dijawab pada merupakan google adwords certified partner. PT.
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.