stedsans is a package capable of doing geospatial analyses from text.
This repository is for an exam project for the course Spatial Analytics at Aarhus University during the spring of 2021.
It is made by Jakob Grøhn Damgaard and Malte Højmark-Bertelsen.
For a demonstration, we heavily suggest you to use Google Colab:
The purpose of it is to build a PyPI-package capable of plotting a map of any location in a Danish sentence. To do so we employ the Natural Language Processing (NLP) technique Named Entity Recognition (NER)
NER is a task consisting of finding words in text that constitute a specific entities and tagging them with specific labels. The most common entities are person names (PER), locations (LOC) and organizations (ORG) (Ruder, 2019). The way the named entities are tagged follows a tagging scheme called BIO-tagging, where the different words are separated as either being the beginning (B) of an entity, inside an entity (I), or other (O), meaning that a word is not part of the defined entities. An illustration of the aforementioned entities can be seen in Table 1.
|B-PER||Beginning of person name|
|I-PER||Inside a person name|
|B-LOC||Beginning of location|
|I-LOC||Inside a location|
|B-ORG||Beginning of organization|
|I-ORG||Inside an organization|
To use the code locally, start off by cloning the repository and install Anaconda for your OS. Afterwards create a conda environment and install the requirements.
# From the directory of this repository conda create -n [env_name] python=3.8 # Create conda environment conda activate [env_name] # Activate conda environment pip install -r requirements.txt # Install required packages
geopandasusing the pre-build binaries from Anaconda:
conda install geopandas
Ruder, S. (2019). Neural transfer learning for natural language processing (Doctoral dissertation, NUI Galway).
Release history Release notifications | RSS feed
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Hashes for stedsans-0.0.12-py3-none-any.whl