Word List 'Finished' after 18 Months

After 18 months of intense heads down development I have finally 'finished' my list of 5000 English words (6M). I picked 5000 words to describe to my computer so that my computer has a chance of understanding them.

I developed attributes for each word using my samod approach. A samod is a list of characteristics which when ordered for a particular item describe that item. In this case the samod I used to describe words was: A-area, B-biological, C-change, D-directive, E-express, F-feeling, G-good, H-human, I-incident, J-jump(movement), K-kind(category), L-language, M-mental, N-number, O-object, P-physical, Q-quality, R-resource, S-system/relationship, T-temporal, U-ugly/bad. Each word in the list has been described using the samod by ordering the list differently for each by showing the order using the letters.

The file containing the list of words has the following sets of columns:

• 24 columns cover the word and its letters. Colors are only there to help me manage the list, and would not be used by the computer.
• Ten columns cover special information related to using the words.
• Three columns for non-English equivalents.
• Seven columns for another samod related to parts of speech (see page 3 for the samod).
• Seven columns for another samod having to do with cognitive types.
• Two columns to describe when the words were probably 'invented' (after all I am working on an artificial creativity time travel game).
• Six columns for roots and such.
• Six columns for word frequency in English.
• Twelve columns for various simple characteristics.
• Five columns for my original 16 characteristics.

The file also contains the following sheets: Sheet 1 contains the word list as shown above. Sheets 3 and 6 describe the characteristics in detail, and future directions for the program. Sheets 2 and 4 list words that I decided to leave out of the word list. Sheet 5 describes my old status tracking system.

The idea of this list was to model something bigger and more complex than I had modeled before, and hope that this could allow players and characters in the artificial world to talk at some level. If I can find somewhere else on the web where someone else has come up with attributes for words I can merge them in and improve the list. The idea of artificial creativity is to come up with models that allow a computer to use them and create things. This is another reason for this model. Finally, I just wanted to try it and see how far I can get.

[0] comments (449 views) |  [0] Trackbacks   [0] Pingbacks

A Data Model for a System to Match Employees With Jobs Using the Cognitive Samod

Last week I intriduced the cognitive samod as a way to match people up with jobs. This week, I am providing a data model that will allow companies (groups) to do this. Here is is:

[0] comments (427 views) |  [0] Trackbacks   [0] Pingbacks

Introducing the Cognitive Samod, a Data Structure That Can Make Hiring Cheaper and More Effective

Every company is competing for the same small pool of candidates[1]. This is expensive for the companies that get them, and wasteful for the companies that don’t. Using the samod[2] idea I have developed for the construction of artificially creative systems, I have designed a samod that can solve the problem.

I've designed a system which would allow you to be more flexible in who you can hire and avoid intense competition for the same small pool of candidates. What is requires is that you hire and use people in a different way. You hire valuable people and then break jobs into job functions and then assign individual job functions rather than entire jobs.

I've designed a system that will allow you to treat every candidate as an individual rather than as a 'yes/no status for a position'. You could be doing half the effort hiring and get twice the value. You would be looking through fewer candidates to fill more positions and you would be making those positions and thus your company organization more flexible. Instead of getting some moderately good matches and a few mistakes, you would be getting very good to excellent matches with fewer mistakes.

My system will allow people to hire people and fit them very fast into the organization rather than hiring candidates for slots and then only fitting them in after months or years or never fitting them in very well at all. Presently 1/3 of positions are poorly matched to the employees filling them. Probably only about 10% are excellently matched. Lets reverse those figures.

My system works by doing a fairly straight-forward cognitive analysis of each candidate and each employee, and a similarly straight-forward analysis of the cognitive needs for each job function using the same cognitive characteristic set. Thus by comparing apples to apples a clear identification of which employee fits which job function comes about. Then, once all employees are assigned to job functions, the remaining poorly filled or non-filled job functions provide the information you need to identify which candidates you need to hire. You end up with a highly cognitively diverse company that is highly productive.

The only problem you have left is how do you manage a highly productive and invigorated workforce. The answer is to apply the same principles to highly empower and invigorate the management. The only thing you have to worry about then is where to build the new buildings as your company grows and grows.

My system is based on the following samod[2] of these cognitive characteristics:
A-Appearance Intelligence, B-Bravery, C-Creativity, D-Deductive Reasoning, E-Emotional IQ, F-Financial IQ, G-Growth, H-Presence, I-Inductive Reasoning, J-Physical Intelligence, K-Kinetic Intelligence, L-Language Intelligence, M-Mechanical Intelligence, N-Navigation, O-Organizational IQ, P-Practicality, Q-Recognition/Matching, R-Reason, S-Service, T-Teamwork, U-Office Work

Each candidate, employee, job function, and business need is evaluated based on these characteristics. Then an AI matching proram artificially creates a number of good possibilities for job function sharing, or switching, or employee reassignment. Those suggestions that fit with other criteria are done, and the employees and thus the company becomes more productive.



[1] I’ve seen this in my own field where every company seems to want a heads down 90% code developer.

[2] - A samod is a list used to profile things by arranging the list
in a different order for each profile. Samod is an old English word meaning
'together'. The connotation is that the ordered list is itself a
thing, created by the combination, and is more than the sum of its parts

[0] comments (470 views) |  [0] Trackbacks   [0] Pingbacks