WordSeer is a text analysis environment that combines visualization, information retrieval, sensemaking and natural language processing to make the contents of text navigable, accessible, and useful. You can run WordSeer on your local machine or on a shared server, and you access it via your modern web browser of choice (Google Chrome works best, though).
If you're a scholar who's interested in using Wordseer for your research, read on. If you want to learn more about how to contribute to WordSeer's ongoing design and development, take a look at our Guidelines for Contributors.
You can find much more information on WordSeer, including demo videos, use case studies, and background research, at wordseer.berkeley.edu.
Be prepared: you may have to do a little command-line voodoo to install and launch WordSeer! But once it is up and running, you will access it through a web browser.
The latest release of WordSeer is available as a .zip or .tar.gz file from this GitHub repository here. Download this file and extract its contents wherever you want; WordSeer will keep all the files it needs to run in this folder.
Before attempting to install WordSeer, make sure your computer has the following software installed:
- Python 2.7 (Python 3.x is NOT supported at this time)
- Java Development Kit 8
Double-click Install Wordseer (Mac) in the folder where you extracted WordSeer and follow the prompts.
You may need administrator privileges to complete some parts of the installation.
Double click on install.bat in the folder where you extracted WordSeer and follow the prompts.
You may need administrator privileges to complete some parts of the installation.
Open the extracted WordSeer folder in your command-line terminal.
From this directory, start the install process by entering the following command:
./install.py -i
This will launch the interactive installer which will guide you through the simple installation process with a series of questions. You may need administrator privileges to complete some parts of the installation.
If you know what you want, run install.py -h to view known console flags.
After installation has completed, you are ready to run WordSeer.
Double-click Wordseer (Mac).app.
You will see a console window with an IP address. The launcher will open this address in your default web browser, but if you close that window and need to return to WordSeer, use this address.
To quit Wordseer, close the console window.
Double-click wordseer.bat.
You will see a console window with an IP address. Navigate to that address in your browser and you should see the WordSeer welcome screen.
If you installed WordSeer in a virtual environment, be sure to activate it first:
source venv/bin/activate
Then launch WordSeer with the following command:
./wordseer.py
You will see a console window with an IP address. Navigate to that address in your browser and you should see the WordSeer welcome screen.
You will be asked to register and log in to use WordSeer; this allows multiple users to collaborate on analysis projects, and the account information you enter will only be stored locally on your computer. There is no mothership for WordSeer to talk to, and we don't collect any information from you.
Once you've logged in, you can create new Projects, which are self-contained collections of documents that you want to analyze. To explore your documents in WordSeer, you must first upload and process them. WordSeer only accepts documents in XML format at this point.
If you want to test out WordSeer before loading your own collection, we have included several
example collections with this download; they are located in the tests/data/ directory. Each
folder in this directory contains a number of .xml files and one .json file. To
explore a demo collection, create a new Project, upload the XML files from your chosen directory
in the "Documents" tab, and upload the JSON file under the "Structure Maps" tab. Then click the
"Process Documents" button (you may need to reload the page after uploading to activate this button)
and choose a structure map from the list that appears (there should be only one, and it should be
the one you just uploaded).
IMPORTANT NOTE: this beta release of WordSeer cannot handle large collections very well. The exact definition of "large" depends on how powerful your machine is, but as a rule of thumb: two or three Shakespeare plays should be fine, but the Complete Works will give you problems. (We could use help fixing this shortcoming: if you or a collaborator think you can help, please see our Guidelines for Contributors and get in touch.)
Take note: processing can be SLOW. The good news, though, is that you only need to do it once. Depending on the number, length, and complexity of the documents in your project, it may take anywhere from 30 seconds to several hours. Processing runs in the background and you can close the browser window and continue using your computer normally while it is in progress, but it will be paused and may encounter errors if your computer goes to sleep. If you close the terminal window where WordSeer was launched from, or shut off your computer, the process will be interrupted and you will have to start over with a new project. The "Processing Logs" tab on the project page will keep you updated on WordSeer's progress; when processing is complete, the "Explore data" button will become active; clicking it launches the WordSeer text analysis environment for that project.
You can learn more about the analysis capabilities of WordSeer and view some video demos and tutorials at our website.
If you've gotten this far, you probably have your own collection of documents that you want to analyze. Good news: as of version 4.0, WordSeer includes a Structure Mapper tool that lets you tell WordSeer what information you want to extract from the XML documents you upload. To access it, upload your documents to a new Project and click the "Map document structure" button. The interface includes detailed instructions for using it; in addition, here are a few things to keep in mind:
- You can make as many Structure Maps as you want, but in the end you will have to choose just one to process the entire collection with; this means all the documents in your Project must follow the same schema.
- WordSeer's Structure Mapper uses just one of your documents at a time as a model for the entire collection; make sure you choose one that is as complete as possible in representing the schema of your documents.
The Structure Mapper creates a JSON file just like the ones that come bundled with our example projects. Once you have created one, you can process and analyze the project in the same manner as described above.
WordSeer is open-source beta software, so you are likely to encounter bugs as you use it. Feel free to report them on our GitHub Issues page (but please check to see if someone else already has, so you don't create a duplicate report).
This software is released as is and is not supported. It was part of a research project that is now complete. We hope that people find it useful and take it upon themselves to improve it.
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.