Đã Hủy

Build a robust international address parser.

We are going to parse freeform street/postal address out of the text, and into components.

[login to view URL]

There are many sample tools and services for address parsing, but we need to retrain on our own address database because they are not accurate sometimes on the specific addresses.

For example, Libpostal is an open-source library to parse addresses, training with data from OpenStreetMap, OpenAddresses, and OpenCage.

[login to view URL]

We want to retrain the Libpostal on our own address database.

You can suggest a better open-source than Libpostal, but we need the pipeline on how to retrain the service on our own database and how to validate the trained model including how to annotate the address strings.

Kĩ năng: Machine Learning (ML), Python, Neural Networks, Deep Learning

Về Bên Thuê:
( 0 nhận xét ) Dublin, United States

ID dự án: #23375929

1 freelancer đang chào giá trung bình $17 cho công việc này


Hi, I'm reaching out about I have been working on Machine Learning, Pattern Recognition, Event Tracking and Natural Language Processing using Python. Below are areas where I am strong and experienced. N.L.P: Get Thêm

$17 USD / giờ
(12 Nhận xét)