Professional skills analysis

A professional skills analysis is done by comparing your birth chart with hundreds of birth charts from celebrities belonging to a professional category of your choice. This process is done with AstroCat (short for "astrological categorisation), a neural network, which has been designed and trained for the purpose. NB! This is a unique product from Sky-Service - nobody else in the astrological community has managed to accomplish, what AstroCat has accomplished!

A professional skills analysis is comparable with a DNA test rather than traditional astrology. AstroCat dissolves a birth chart into 36,360 floating point values by use of harmonics before processing it. That's almost 6 times as many as the number of DNA in the human genome, ref.: "Genome size 3,100 Mbp (mega-basepairs) per haploid genome 6,200 Mbp total (diploid)." (Quote: Wikipedia)

Precision: In tests AstroCat has proven to be 95% correct, when it comes to detecting the 50% best athletes in a population - and 75% correct, when it comes to detecting the 5% best athletes in a population. The difference in precision is caused by the content of the training set: If the training set consisted of the 5% best athletes only, then the precision would be increased. But since AstroCat needs a relatively high number of athletes' birth charts in the training set in order to be able to generalize, I have been forced to include less eminent athletes as well in the training set. (You can substitute the word "athletes" with any other category, because the final precision is very much the same with every category.)

Figure 1 illustrates the output of AstroCat. The first bell-shaped curve represents AstroCat's output for the population in general, when the birth charts are compared with the athletes' birth charts in the training set. The second bell-shaped curve represents the output for the eminent athletes in the test set. As you can see, there is always going to be some "outliers", who don't fit in, simply because the AstroCat models reality - it does not represent an actual copy of the reality, because such a precision would be impossible. A model is always a simplification of real conditions.

Figure 1

Below 2 lists of professional categories are presented, one for each gender. More professional categories can (and will) be added, provided that charts from the selected professional categories exist in the available data collections.

Females, professions

  • Actress
  • Astrologer
  • Athlete (except for martial arts, motorsport, mountain climbing and teamsport, which are not included yet)
  • Artist (like e.g. photographer, painter or sculptor)
  • Author of fiction (playwrights and poets are not included yet)
  • Lawyer
  • Photo model
  • Physician
  • Politician
  • Singer-songwriter - A separate category with singers, who are not songwriters, is expected to be created in the future.
  • Spy or secret agent - This category, which includes resistence fighters from 2nd world war, was researched out of curiosity, since I don't expect it to have much commercial value.
  • Top executives

At the moment (August 2021) I am collecting data for the next category: Female physicians.

Before AstroCat there was AstroTwin. AstroCat has a higher precision than AstroTwin, when it comes to categorization, but AstroTwin is still the best tool for type casting. The transition from AstroTwin to AstroCat for categorization means that all the below listed categories are under reconstruction. It takes approximately 1 month to train AstroCat for 1 category. The training includes testing the precision of AstroCat, and it is the tests, which are very time consuming. If you wish to order a professional skills analysis, which haven't been reconstructed yet, then order it. Then I shall train AstroCat for the given category, so that I can deliver within 1-2 months. Anything is possible in this area, provided the necessary celebrity birth charts are available in sufficiently large quantities.

  • Author, non-fiction
  • Comedian
  • Composer
  • Dancer
  • Designer
  • Film / Theater director
  • Health worker (specify, e.g. physician, nurse, dentist, etc.)
  • Journalist
  • Musician
  • Organisation leader (e.g. a manager of a school, a hospital, a non-profit organisation, etc.)
  • Producer
  • Psychologist / psychiatrist
  • Religious leader
  • Secretary
  • Self-employed / entrepreneur
  • Sex symbol
  • Singer
  • Social worker
  • Teacher / educator
  • Team sport
  • TV/Radio host

 

Males professions

Only one category is ready, namely:

  • Top executives

It means that the following categories are still waiting to be processed:

  • Actor
  • Advertising
  • Artist
  • Astrologer
  • Author, fiction (specify, e.g. novels, poetry, theater/movie manus, children's litterature)
  • Author, non-fiction
  • Comedian
  • Composer
  • Dancer
  • Designer
  • Engineer / Technician
  • Film / Theater director
  • Health worker (specify, e.g. physician, nurse, dentist, etc.)
  • Journalist
  • Lawyer / Judge
  • Military officer
  • Musician
  • Organisation leader (e.g. a manager of a school, a hospital, a non-profit organisation, etc.)
  • Policeman
  • Politician
  • Producer
  • Programmer
  • Psychologist / psychiatrist
  • Religious leader
  • Sailor
  • Scientist (specify, e.g. mathematician, historian, chemist, etc.)
  • Self-employed / entrepreneur
  • Singer
  • Solo sport (specify, e.g. tennis player, swimmer, track & field, etc.)
  • Teacher / educator
  • Team sport
  • TV/Radio host

90%-95% of an astrologer's clients are women, which is why I have started the reconstruction with the women's categories. After all, it is the people, who pay, who get to decide what research is carried out first. However, if you are male, and you wish to order a professional skills analysis, then order it, and then I can deliver within 1-2 months. Anything is possible in this area, provided the necessary celebrity birth charts are available in sufficiently large quantities.

Time budget for new categories

The process of establishing a new category has become quite set, and it looks like this:

  1. Collecting data: Variable. Sometimes the celebrity birth charts are easily available, at other times some data searching in available collections is necessary.
  2. Rectifications: 3 weeks or 21 work days. However, sometimes I already have the necessary rectified data from a given category, which is then characterized as a "low hanging fruit, easy to pick". But if for instance I wish to do the category "female mathematicians", then I'd have to start from scratch. In some cases like e.g. female authors of childrens' books I have already rectified approximately 50% of the birth charts.
  3. Setting up and running the neural network on a category: 10 working days.

This time buduget assumes that everything runs smoothly and no delays occur.

Precision

The neural network's precision for a given category seems to depend on the following 2 factors mainly:

  • Specialization
  • Competition

For instance, the neural network has a relatively low precision with the politicians compared to other professional categories. I can probably increase the specialization by making a new category containing centre-left politicians only or the right-wing conservative politicians only. Or I can increase the competition by including cabinet members only in a new category.

For these reasons it seems to be futile to create very broad categories like e.g. "scientists". Instead it would be more rewarding to create categories with specific scientific disciplines like e.g. metheorology, history, chemistry or whatever. However, by doing so I risk running into a problem with lack of data for a given category.

The scientific method

The scientific method has five basic steps, plus one feedback step:

  1. Make an observation.
  2. Ask a question.
  3. Form a hypothesis, or testable explanation.
  4. Make a prediction based on the hypothesis.
  5. Test the prediction.
  6. Iterate: Use the results to make new hypotheses or predictions.

Ad 1. Make an observation.

Many observations were made:

My basic observation in this context was that astrology cookbook interpretations on average are 65% correct, which is more than can be expected by mere chance. The best astrological cookbook I have tested so far was 75% correct.

Michel Gauquelin's astrological statistics for birth charts belonging to various professions added to the puzzle.

Another observation I made early in my astrological career was that chart rectification does matter. Once a birth chart has been properly rectified, it is possible to make fairly accurate predictions in e.g. a one-year-prognosis for that individual.

Ad 2. Ask a question.

My question was: Is it possible to turn the above observations into something useful by processing rectified charts belonging to eminent professionals through a neural network?

Ad 3. Form a hypothesis, or testable explanation.

You could say that the various neural network designs I have experimented with since 1997 were expressions of hypotheses. The tests were expensive in time and made the entire process rather slow, but December 23rd 2020 I succeeded with one professional category: Female athletes.

Ad 4. Make a prediction based on the hypothesis.

It's easy to blind test the result:Provide me with the necessary data, i.e. birth date, birth time, birth place plus information (headline: what happened?, date and place) for 6 major events in that person's life. Then I shall rectify the birth chart and process it through AstroCat and present you with the result, which is a number on a scale signifying the likelihood that this person belongs to the category or not.

Ad 5. Test the prediction.

My own tests indicate that AstroCat is precise enough to be useful.

Ad 6. Use the results to make new hypotheses or predictions.

So far I have iterated with 9 different professional categories - and in the current moment (May 20th 2021), I am processing the 10th professional category, which is female singer-songwriters.