Get Started on Alpha release Citizen.chat
Desktop only app for the moment. On the phone it does work, although UI and UX is not optimized. Its Alpha, please bear with us, or even better, let us know what you think!
Last updated
Desktop only app for the moment. On the phone it does work, although UI and UX is not optimized. Its Alpha, please bear with us, or even better, let us know what you think!
Last updated
You can securely login hassle free with a "magic" link that will be sent to your email.
Magic link secure login.
Go to the website citizen.chat and use the blue "Pre-register for the Alpha" button that serves for login.
Open your inbox, click on the button in your email
enter citizen.chat.
fill in the email adress with which you have registered previously.
If you want the latest updates, or have new permissions, make sure to logout previous session, and refresh before login in again.
Go to Library menu item
In top right corner Enter the Name for your dataset.
Choose carefully, as of this moment there is NO WAY to edit the name later.
In each dataset you can upload Documents and upload data from a Website (URL)
Document types:
pdf, doc, docx .eml , .html , .json , .md , .msg , .rst , .rtf , .txt , .xml , .csv , .docx , .epub , .odt , .ppt , .pptx , .tsv, .xlsx, .ZIP
Images with text:
.png, .jpg, .pdf
You can add one document or select several ones.
with Js - With Javascript, Its a workaround for if pages are blocked from indexing because of a javascript firewallthis is useful of there are banners or other blockers on the page. If a website doesn't get the full data, try activating > with Js.
HiRes - Hi Resolution index. This is mostly for urls with pdf's or images or tables, that contain text. Scanning in HiResolution will get more complete data.
Multipage - Multipage Index, If you want to index all pages from a specific domain.
Skip Files - Skip files and images of websites while indexing. Only index text content.
Parents - Index the Parent pages. If you enter the URL of a specific page that is nested in a tree of Webpages, you activate Parents to index also the webpages higher in the webtree.
Outside - Index URL's of links that are outside of the given URL domain. ALERT, depending on the depth you put this setting can increase the number of webpages that will be indexed exponentially.
Depth - Depth of link tree. This will apply if you have activated "Outside". The number here will indicate what is the level of the links tree that will be included in the index.
For example. If you have a blog post with references mentioned It will also index the links that go outside of the URL domain, and index 1 level of linked pages. If you spice up this number to for example 3, it will index upto 3 levels deep. Which means it will repeat this action for the second level URL's it lands on on so on. Which depending on the URL you land on could go out of control pretty fast.
When you start to upload content, the Status opens up by default.
It contains information about:
Task Origin:
CreatedAt: date and time of initiation of the upload.
If it is a single or multiple documents.
The total number of Documents structured extraction + Progress bar
The total number of the Documents embeddings + Progress bar
After you see the same total number is shown for the extractions as well as embeddings, you Refresh your browser to see the documents reflected in the UI.
You can share the dataset you have created. This will result in 2 things:
The dataset will become available in the shared Library. (Refresh browser after sharing )
Others can ask Questions on your dataset.
What is will NOT do:
Disclose and view the documents that you have uploaded in your dataset.
Give permission to other users to upload documents or websites in your dataset.
