27. WIKIPEDIA:
A scientist is a person engaging in a systematic
activity to acquire knowledge that describes and
predicts the natural world.
28. WIKIPEDIA:
Data
A scientist is a person engaging in a systematic
activity to acquire knowledge that describes and
predicts the natural world.
29. WIKIPEDIA:
Data
A scientist is a person engaging in a systematic
activity to acquire knowledge that describes and
predicts the natural world.→ Data
30. WIKIPEDIA:
Data
A scientist is a person engaging in a systematic
activity to acquire knowledge that describes and
predicts the natural world.→ Data
Data: a set of values of qualitative or
quantitative variables.
31. WIKIPEDIA:
Data
A scientist is a person engaging in a systematic
activity to acquire knowledge that describes and
predicts the natural world.→ Data
Data: a set of values of qualitative or
quantitative variables.
32. WIKIPEDIA:
Data
A scientist is a person engaging in a systematic
activity to acquire knowledge that describes and
predicts the natural world.→ Data
Data: a set of values of qualitative or
quantitative variables.
59. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
60. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
61. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
62. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
63. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
- Cross-section VS Time-Series VS Panel
64. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
- Cross-section VS Time-Series VS Panel
- Prediction: Estimation VS Forecasting
65. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
- Cross-section VS Time-Series VS Panel
- Prediction: Estimation VS Forecasting
- Improvement: Accuracy VS Insight
66. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
- Cross-section VS Time-Series VS Panel
- Prediction: Estimation VS Forecasting
- Improvement: Accuracy VS Insight
- Modeling.
67. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
- Cross-section VS Time-Series VS Panel
- Prediction: Estimation VS Forecasting
- Improvement: Accuracy VS Insight
- Modeling.
- Expertise
68. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
- Cross-section VS Time-Series VS Panel
- Prediction: Estimation VS Forecasting
- Improvement: Accuracy VS Insight
- Modeling.
- Expertise
- Econometric
69. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
- Cross-section VS Time-Series VS Panel
- Prediction: Estimation VS Forecasting
- Improvement: Accuracy VS Insight
- Modeling.
- Expertise
- Econometric
- AI
70. Now assume that you have a
cleansed big data set...
- Describe the data using visualization or other appropriate
measurements.
- Define the problem.
- Supervised VS Unsupervised
- Balanced VS Unbalanced
- Cross-section VS Time-Series VS Panel
- Prediction: Estimation VS Forecasting
- Improvement: Accuracy VS Insight
- Modeling.
- Expertise
- Econometric
- AI
- Hybrid
71. Data
Validation set
Training set
Test set
Train
classifier
Homogeneous
ensemble
algorithm
Individual
classification
algorithm
Apply
model
Classification
models
Apply
model
Test set
prediction
Train
classifier
Ensemble model
Validation set
predictions
Apply
model
Heterogeneous
ensemble
algorithm
Features
Selection
Clustering
Estimated
Value
STATISTICAL LEARNING FLOWCHART
PLIZ, OJO NGE-LIB!!
73. Description Value
ROW_ID Row ID NUMERIC
MAIN_PARTNER Nomor referral ID dari Astra World (AWO) NUMERIC
FRAME_NO Nomor rangka motor yang dipunyai customer TEXT
CUST_ID Nomor ID customer yang didapat dari KTP/SIM TEXT
SALES_DATE Tangga sepeda motor honda dibeli DATE (YYYY-MM-DD HH:MM:SS)
KODE_MESIN Tiap tipe motor mempunyai kode mesin yang berbeda dengan tipe motor yang lain 75 NOMINAL {JF81E, ...}
SEQUENCE_MESIN Sequence dari kode mesin NUMERIC
VARIAN_MOTOR Varian motor yang dipunyai customer 76 NOMINAL {ALL NEW VARIO, …}
COLOR Warna motor yang dipunyai customer 73 NOMINAL {HITAM, …}
KODE_CUSTOMER Tipe customer {INDIVIDUAL, COLLECTIVE, GROUP, JOINT PROMO}
JENIS_KELAMIN Jenis kelamin customer {LAKI-LAKI, PEREMPUAN}
TANGGAL_LAHIR Bulan dan tahun lahir customer DATE (MM/YYYY)
KELURAHAN_SURAT Kelurahan surat menyurat customer 1251 NOMINAL {KETEWEL, …}
KECAMATAN_SURAT Kecamatan surat menyurat customer 120 NOMINAL {SUKAWATI, …}
KOTA_SURAT Kota surat menyurat customer 30 NOMINAL {KAB. GIANYAR, …}
KODE_POS Kode pos surat menyurat customer NUMERIC
PROPINSI Propinsi surat menyurat customer 8 NOMINAL {BALI, …}
STATUS_RUMAH Status rumah customer {RUMAH SENDIRI, RUMAH SEWA, RUMAH ORANG TUA/KELUARGA}
JENIS_PENJUALAN_STNK Jenis penjualan saat keluar faktur (bener-benar terjual) {CASH, CREDIT}
JENIS_PENJUALAN_SSU Jenis penjualan ini saat deal, bisa berubah saat transaksi {CASH, CREDIT}
NAMA_LEASING_COMPANY Nama leasing company yang menangani cicilan customer TEXT
BESAR_DP Besar DP yang diberikan customer TEXT
BESAR_CICILAN Besar cicilan per bulan NUMERIC
LAMA_CICILAN Lama cicilan sampai lunas (bulan) NUMERIC
AGAMA Agama customer {HINDU, KRISTEN, ISLAM, KATOLIK, LAIN-LAIN, BUDHA}
PEKERJAAN Pekerjaan customer 16 NOMINAL {PEGAWAI SWASTA, …}
PENGELUARAN Pengeluaran customer per bulan {1,2,3,4,5,6,7}
PENDIDIKAN Pendidikan terakhir customer {SLTA/SMU, AKADEMI/DIPLOMA, TIDAK TAMAT SD, SD, SLTP/SMP, SARJANA, PASCA SARJANA}
NO_HP Nomor handphone customer TEXT
STATUS_NOMOR_HP Tipe kartu handphone customer {PRABAYAR, PASCABAYAR}
NO_TLP Nomor telepon customer TEXT
KEBERSEDIAAN DIHUBUNGI Kebersediaan customer untuk dihubungi lagi di masa depan {YES, NO}
MERK_MOTOR_SBLMNYA Merk motor yang dipunyai customer sebelumnya {HONDA, YAMAHA, SUZUKI, BELUM PERNAH MEMILIKI, KAWASAKI, MOTOR LAIN}
TYPE_MOTOR_SBLMNYA Tipe motor yang dipunyai customer sebelumnya {AT AUTOMATIC, CUB BEBEK, SPORT, BELUM PERNAH MEMILIKI}
SMH_DIGUNAKAN_UNTUK Tujuan dibelinya sepeda motor {LAIN-LAIN, KEBUTUHAN KELUARGA, KE SEKOLAH/ KE KAMPUS, BERDAGANG, PEMAKAIAN JARAK DEKAT, REKREASI / OLAH RAGA, BEKERJA}
YG_MENGGUNAKAN_SMH Orang yang akan menggunakan sepeda motor yang dibeli {ANAK, LAIN-LAIN, PASANGAN SUAMI ATAU ISTRI, SAYA SENDIRI}
MD Kode Main Dealer yang membawahi dealer tempat customer membeli sepeda motor Honda {N01}
DEALER_CODE Kode dealer tempat customer membeli sepeda motor Honda 77 NOMINAL {06877, …}
KODE_SALES_PERSON Kode sales person yang menjual sepeda motor Honda ke customer 1718 NOMINAL {218595, …}
TGL_MASUK_DATA Tanggal masuk ke AHM dari MD DATE (YYYY-MM-DD HH:MM:SS)
STATUS_VALIDASI Validasi dari MD untuk menandakan apakah baris data CDB terkait sudah divalidasi kebenarannya atau belum {1,2}
UPLOADED_ON Tanggal masuk ke AWO dari AHM DATE (YYYY-MM-DD HH:MM:SS)
96. Description Value
ROW_ID Row ID NUMERIC
MAIN_PARTNER Nomor referral ID dari Astra World (AWO) NUMERIC
FRAME_NO Nomor rangka motor yang dipunyai customer TEXT
CUST_ID Nomor ID customer yang didapat dari KTP/SIM TEXT
SALES_DATE Tangga sepeda motor honda dibeli DATE (YYYY-MM-DD HH:MM:SS)
KODE_MESIN Tiap tipe motor mempunyai kode mesin yang berbeda dengan tipe motor yang lain 75 NOMINAL {JF81E, ...}
SEQUENCE_MESIN Sequence dari kode mesin NUMERIC
VARIAN_MOTOR Varian motor yang dipunyai customer 76 NOMINAL {ALL NEW VARIO, …}
COLOR Warna motor yang dipunyai customer 73 NOMINAL {HITAM, …}
KODE_CUSTOMER Tipe customer {INDIVIDUAL, COLLECTIVE, GROUP, JOINT PROMO}
JENIS_KELAMIN Jenis kelamin customer {LAKI-LAKI, PEREMPUAN}
TANGGAL_LAHIR Bulan dan tahun lahir customer DATE (MM/YYYY)
KELURAHAN_SURAT Kelurahan surat menyurat customer 1251 NOMINAL {KETEWEL, …}
KECAMATAN_SURAT Kecamatan surat menyurat customer 120 NOMINAL {SUKAWATI, …}
KOTA_SURAT Kota surat menyurat customer 30 NOMINAL {KAB. GIANYAR, …}
KODE_POS Kode pos surat menyurat customer NUMERIC
PROPINSI Propinsi surat menyurat customer 8 NOMINAL {BALI, …}
STATUS_RUMAH Status rumah customer {RUMAH SENDIRI, RUMAH SEWA, RUMAH ORANG TUA/KELUARGA}
JENIS_PENJUALAN_STNK Jenis penjualan saat keluar faktur (bener-benar terjual) {CASH, CREDIT}
JENIS_PENJUALAN_SSU Jenis penjualan ini saat deal, bisa berubah saat transaksi {CASH, CREDIT}
NAMA_LEASING_COMPANY Nama leasing company yang menangani cicilan customer TEXT
BESAR_DP Besar DP yang diberikan customer TEXT
BESAR_CICILAN Besar cicilan per bulan NUMERIC
LAMA_CICILAN Lama cicilan sampai lunas (bulan) NUMERIC
AGAMA Agama customer {HINDU, KRISTEN, ISLAM, KATOLIK, LAIN-LAIN, BUDHA}
PEKERJAAN Pekerjaan customer 16 NOMINAL {PEGAWAI SWASTA, …}
PENGELUARAN Pengeluaran customer per bulan {1,2,3,4,5,6,7}
PENDIDIKAN Pendidikan terakhir customer {SLTA/SMU, AKADEMI/DIPLOMA, TIDAK TAMAT SD, SD, SLTP/SMP, SARJANA, PASCA SARJANA}
NO_HP Nomor handphone customer TEXT
STATUS_NOMOR_HP Tipe kartu handphone customer {PRABAYAR, PASCABAYAR}
NO_TLP Nomor telepon customer TEXT
KEBERSEDIAAN DIHUBUNGI Kebersediaan customer untuk dihubungi lagi di masa depan {YES, NO}
MERK_MOTOR_SBLMNYA Merk motor yang dipunyai customer sebelumnya {HONDA, YAMAHA, SUZUKI, BELUM PERNAH MEMILIKI, KAWASAKI, MOTOR LAIN}
TYPE_MOTOR_SBLMNYA Tipe motor yang dipunyai customer sebelumnya {AT AUTOMATIC, CUB BEBEK, SPORT, BELUM PERNAH MEMILIKI}
SMH_DIGUNAKAN_UNTUK Tujuan dibelinya sepeda motor {LAIN-LAIN, KEBUTUHAN KELUARGA, KE SEKOLAH/ KE KAMPUS, BERDAGANG, PEMAKAIAN JARAK DEKAT, REKREASI / OLAH RAGA, BEKERJA}
YG_MENGGUNAKAN_SMH Orang yang akan menggunakan sepeda motor yang dibeli {ANAK, LAIN-LAIN, PASANGAN SUAMI ATAU ISTRI, SAYA SENDIRI}
MD Kode Main Dealer yang membawahi dealer tempat customer membeli sepeda motor Honda {N01}
DEALER_CODE Kode dealer tempat customer membeli sepeda motor Honda 77 NOMINAL {06877, …}
KODE_SALES_PERSON Kode sales person yang menjual sepeda motor Honda ke customer 1718 NOMINAL {218595, …}
TGL_MASUK_DATA Tanggal masuk ke AHM dari MD DATE (YYYY-MM-DD HH:MM:SS)
STATUS_VALIDASI Validasi dari MD untuk menandakan apakah baris data CDB terkait sudah divalidasi kebenarannya atau belum {1,2}
UPLOADED_ON Tanggal masuk ke AWO dari AHM DATE (YYYY-MM-DD HH:MM:SS)