A Data Science Central Community

To be able to explain making assumptions, we need to understand (the difference with) drawing conclusions first:

• Conclusions are drawn straight ahead - top-down - like in: Given "*John is a father.*" and "*A father is a man.*". Generated conclusion: "*John is a man.*" (see Part 2);

• Assumptions however, are made bottom-up, by which the validity is not secured. So, we should add a word that expresses the uncertainty of the assumption.

Under certain conditions, two types of assumptions can be generated autonomously:**1) Generalized Assumptions:**

• Given: "*A parent is a father or (a) mother.*"

A rule of intelligence contained within grammar: Conjunction "*or*" can be used to dissected the above sentence into two singular definitions:

• "*A father is a parent.*"

• "*A mother is a parent.*"

Now take one of the dissected sentences and another sentence:

• "*A father is a parent.*";

• "*John is a father.*" or "*John is the father of Pete.*"

The common word in both sentences is "*father*". At this point we can draw a *Specification Substitution Conclusion* (see Part 2) from the original sentence "*A parent is a father or (a) mother.*" and the sentence, "*John is the father of Pete.*". We can now conclude - actually assume - that "*John is a parent (of Pete).*".

Below we will learn how the uncertainty of the assumption is expressed.

(More detailed conditions: http://mafait.org/en/theory_2_4_1/)**2) Possessive Conditional Specification Assumptions:**

• Given: "*A family has parents and children.*" and "*John is a parent of Pete.*";

The first sentence has a prior term of the possessive verb "*to have*". And its conjunction "*and*" - another rule of intelligence - indicates that both specifications "*parents*" and "*children*" are necessary to validate the definition "*family*".

In the second sentence, we can read a relational specification: "*John has a parent(al) relation with Pete.*". However, the reverse relationship from Pete to John is unknown, in which case we may not make any conclusions about Pete, although we would like to claim that:

• "*Pete is a child of John.*"

• "*John has a child (named Pete).*"

Both we call a *Possessive Conditional Specification Assumption*.

(More detailed conditions: http://mafait.org/en/theory_2_4_2/)**Expressing the level of uncertainty of an assumption:**

• The "assumption distance" can be calculated by counting the number of bottom-up steps in the assumption;

• And we can translate this number of steps into a word with increasing uncertainty, like: 1 step = "*probably*", 2 steps = "*possibly*", 3 steps = "*may be*", etc;

• When involved knowledge has changed, the uncertainty level of an assumption should be recalculated.**Example 1:** Above we have made the assumption "*John is a parent (of Pete).*" from "*A father is a parent.*" and "*John is the father of Pete.*". Jumping from the bottom word "*father*" to upper word "*parent*" is only one step. So, we can express the uncertainty of this one step as: "*John is probably a parent (of Pete).*".**Example 2:** An assumption might include another assumption. Then the level of uncertainty of the included assumption will add to the number of steps of the assumption to be generated. Known: "*A family has parents and children.*" and from the previous example: "*John is probably a parent of Pete.*". Now an assumption can be generated autonomously: "*John has possibly a child, called Paul.*", whereas the word "*possibly*" expresses the extra assumption step, a double level of uncertainty.

To download the open source implementation: http://mafait.org/en/download/

© 2019 AnalyticBridge.com is a subsidiary and dedicated channel of Data Science Central LLC Powered by

Badges | Report an Issue | Privacy Policy | Terms of Service

**Most Popular Content on DSC**

To not miss this type of content in the future, subscribe to our newsletter.

- Book: Statistics -- New Foundations, Toolbox, and Machine Learning Recipes
- Book: Classification and Regression In a Weekend - With Python
- Book: Applied Stochastic Processes
- Long-range Correlations in Time Series: Modeling, Testing, Case Study
- How to Automatically Determine the Number of Clusters in your Data
- New Machine Learning Cheat Sheet | Old one
- Confidence Intervals Without Pain - With Resampling
- Advanced Machine Learning with Basic Excel
- New Perspectives on Statistical Distributions and Deep Learning
- Fascinating New Results in the Theory of Randomness
- Fast Combinatorial Feature Selection

**Other popular resources**

- Comprehensive Repository of Data Science and ML Resources
- Statistical Concepts Explained in Simple English
- Machine Learning Concepts Explained in One Picture
- 100 Data Science Interview Questions and Answers
- Cheat Sheets | Curated Articles | Search | Jobs | Courses
- Post a Blog | Forum Questions | Books | Salaries | News

**Archives:** 2008-2014 |
2015-2016 |
2017-2019 |
Book 1 |
Book 2 |
More

**Most popular articles**

- Free Book and Resources for DSC Members
- New Perspectives on Statistical Distributions and Deep Learning
- Time series, Growth Modeling and Data Science Wizardy
- Statistical Concepts Explained in Simple English
- Machine Learning Concepts Explained in One Picture
- Comprehensive Repository of Data Science and ML Resources
- Advanced Machine Learning with Basic Excel
- Difference between ML, Data Science, AI, Deep Learning, and Statistics
- Selected Business Analytics, Data Science and ML articles
- How to Automatically Determine the Number of Clusters in your Data
- Fascinating New Results in the Theory of Randomness
- Hire a Data Scientist | Search DSC | Find a Job
- Post a Blog | Forum Questions

## You need to be a member of AnalyticBridge to add comments!

Join AnalyticBridge