Insight Consulting Project: Shipamax Database

Cargo request organization in a real-time updatable, searchable SQL database.

Transcript

1. Callie Bradley
2. 2/13/17 Callie Bradley | Insight 2
3. 2/13/17 Callie Bradley | Insight 3 Container Shipping
4. 2/13/17 Callie Bradley | Insight 4 Container Shipping
5. 2/13/17 Callie Bradley | Insight 5 Container Shipping Tramp Shipping General Cargo . Tankers . Dry Bulk
6. 2/13/17 Callie Bradley | Insight 6 Container Shipping Tramp Shipping General Cargo . Tankers . Dry Bulk Largely low-tech industry...
7. 2/13/17 Callie Bradley | Insight 7 Ship managers have to sift through 1,000+ emails
8. 2/13/17 Callie Bradley | Insight 8 Ship managers have to sift through 1,000+ emails
9. 2/13/17 Callie Bradley | Insight 9 Ship managers have to sift through 1,000+ emails
10. 2/13/17 Callie Bradley | Insight 10 Ship managers have to sift through 1,000+ emails
11. 2/13/17 Callie Bradley | Insight 11 Ship managers have to sift through 1,000+ emails
12. Real-time updating, searchable database of cargo requests 2/13/17 Callie Bradley | Insight 12
13. Real-time updating, searchable database of cargo requests 2/13/17 Callie Bradley | Insight 13 Jakarta
14. Real-time updating, searchable database of cargo requests 2/13/17 Callie Bradley | Insight 14 Jakarta Jakarta, Indonesia (Feb 01 – Jul 31, 2017) Destination Cargo Arrival Date FIOST Commission Shanghai, CN 8000 MT Bulk Iron 24-27 Jul 2.5% Cartagena, IT 2.5 MT Marble 7 days SSHEX 2.5% ...
15. 2/13/17 Callie Bradley | Insight 15
16. 2/13/17 Callie Bradley | Insight 16
17. 2/13/17 Callie Bradley | Insight 17
18. 2/13/17 Callie Bradley | Insight 18
19. 2/13/17 Callie Bradley | Insight 19
20. 2/13/17 Callie Bradley | Insight 20 1. Decision tree request separation (pattern recognition) 20+ pattern dictionary Request separation variations
21. 2/13/17 Callie Bradley | Insight 21 NLTK pre-processing Disparate headers, footers, pattern-distractions Filter out non-request emails Decision tree request separation (pattern recognition) 20+ pattern dictionary
22. 2/13/17 Callie Bradley | Insight 22 Feature engineering 6000+ port location SQL table
 Learns locations Key cargos recognition 
 Date approximations NLTK pre-processing Disparate headers, footers, pattern- distractions Filter out non-request emails Decision tree request separation (pattern recognition) 20+ pattern dictionary
23. 2/13/17 Callie Bradley | Insight 23 Destination Cargo Arrival Date Shanghai, CN 8000 MT Bulk Iron 24-27 Jul Cartagena, IT 2.5 MT Marble 7 days SSHEX ... OUTPUT: SQL DATABASE real-time updating (Python program), searchable
24. 24 Callie Bradley Ph.D. Particle Physics University of OK LHCollider at CERN DESY Postdoc Germany [GeV] pole t m 164 166 168 170 172 174 176 178 180 182 Cross-section [pb] 150 200 250 300 350 MSTW 2008 NNLO MSTW 2008 NNLO uncertainty CT10 NNLO CT10 NNLO uncertainty NNPDF2.3 NNLO NNPDF2.3 NNLO uncertainty -1 = 7 TeV, 4.6 fb s -1 = 8 TeV, 20.3 fb s ATLAS 7 TeV 8 TeV } t vs m Adventure . Build . Yoga . Travel