News Aggregator Service for Astrology (NASA): A Free Research Tool
Welcome. This free tool culls news sites for astrologically actionable data.
Provided for each entry is:
Every day, it gets smarter, and new events are added, so check back often.
Provided for each entry is:
- category (for example, "hurricane")
- event date
- event time
- event place
- keywords (for example, "katrina")
- originating source URL for the data
- URL publication date.
Every day, it gets smarter, and new events are added, so check back often.
Currently, the following separate categories of data being monitored are:
- accident
- announcement
- assassination
- birth
- bitcoin
- coup
- crash
- cyclone
- death
- earthquake
- epidemic
- explosion
- fire
- game
- hospitalization
- hurricane
- IPO
- match
- merger
- miracle
- pandemic
- premiere
- shooting
- tsunami
- volcano
- wedding.
Do you have a favorite research topic that is not included here? Submit a different category for consideration:
Presently, the news publications being monitored are:
- The New York Times
- The Guardian
- Der Spiegel (English edition)
- U. S. News & World Report
- BBC
- Wikipedia Current Events.
Submit a different publication for consideration:
Culling of news data typically happens once or twice daily and is aggregated over the time.
Today's up-to-date news tally for each category is the following:
Today's up-to-date news tally for each category is the following:
Note that this is an automated service. Data are not checked by hand before inclusion. Ymmv.
Acquire new news data.
After clicking on the above link, in the new browser window simply choose the "save" option to download the data to your hard drive. You will then be able to open it by any software that can handle .csv files. For example, you can use R, Python, Excel or LibreOffice to further select keywords and to change the format for inclusion in your particular astrology software. Going to the URL allows you to cross-check the data.
Version release notes for the news aggregator:
0.3.0.3 Feb 18, 2020: Updated code with improvements. Stricter Standards employed where date, time,and location all are required. Removed Beijing Bulletin from sources. Added date collected column. Deleted all prior data,
April 24, 2020: Added "epidemic" and "pandemic" as subjects of interest. April 26, 2020: Added "miracle" as a subject of interest.
0.3.0.2 Nov 26, 2019: Based on community suggestions, "bitcoin" and "announcement" were added to the domain list. Keep them coming.
0.3.0.2 Oct 20, 2019: implemented BERT instead of ELMo for determining contextual sentences for article time, date, and place.
0.3.0.1 Apr 3, 2019: improved algorithm, start of collection
0.3.0.0 Mar 26, 2019: initial commit; results not included in the data
0.3.0.3 Feb 18, 2020: Updated code with improvements. Stricter Standards employed where date, time,and location all are required. Removed Beijing Bulletin from sources. Added date collected column. Deleted all prior data,
April 24, 2020: Added "epidemic" and "pandemic" as subjects of interest. April 26, 2020: Added "miracle" as a subject of interest.
0.3.0.2 Nov 26, 2019: Based on community suggestions, "bitcoin" and "announcement" were added to the domain list. Keep them coming.
0.3.0.2 Oct 20, 2019: implemented BERT instead of ELMo for determining contextual sentences for article time, date, and place.
0.3.0.1 Apr 3, 2019: improved algorithm, start of collection
0.3.0.0 Mar 26, 2019: initial commit; results not included in the data
Future goals:
Improve neural net nlp models in order to reduce false positives in the news aggregator.
Improve neural net nlp models in order to reduce false positives in the news aggregator.
In contrast to new news data, the service is also interested in the much slower job of searching for day, time, and place for historical events for each category, as described in Wikipedia articles. The virtues of this different file are that it is far more comprehensive and the data tend to be of greater accuracy.
The tally so far for each category in the historical Wikipedia file is:
The tally so far for each category in the historical Wikipedia file is:
We are currently in the midst of the active collection process, however. So, check back often for addenda.
Acquire historical Wikipedia data.
After clicking on the above link, in the new browser window simply choose the "save" option to download the data to your hard drive. You will then be able to open it by any software that can handle .csv files. For example, you can use R, Python, Excel or LibreOffice to further select keywords and to change the format for inclusion in your particular astrology software. Going to the given subject in Wikipedia allows you to cross-check the data.
Version release notes for the Wikipedia data:
0.11 Oct 21, 2019: BERT net chosen over ELMo. Service was resumed. All data from version 0.1 was deleted.
0.1: Collection started Mar 31, 2019 and is scheduled to run until early May, 2019, in order to cull for the first time all categories. [Actually, service was put on hiatus in mid April, 2019.]
Version release notes for the Wikipedia data:
0.11 Oct 21, 2019: BERT net chosen over ELMo. Service was resumed. All data from version 0.1 was deleted.
0.1: Collection started Mar 31, 2019 and is scheduled to run until early May, 2019, in order to cull for the first time all categories. [Actually, service was put on hiatus in mid April, 2019.]
A diagram of the complexity of just one of the neural nets used in the NASA research tool
If you would like to read more on the background of this tool, here is a short whitepaper.