1.Abstract:

 

The base technique of a traditional machine learning or artificial intelligence algorithm has a target for learning and the learning technique based on mainly pattern matching, like artificial neural network has a target and for every learning attempt it matches it’s input with the target and adjust the weights to get the target. If we consider genetic algorithm, we have to calculate fitness function and evaluate the function for the target….. And so on.

Thinking simply these algorithms are working / learning based on pattern matching. Or if we think more simply, these techniques are like that some data stored in database, machines are doing nothing but finding, probability calculating, predicting some results based on matching .

When we are using search engines, we just enter some keywords and the search engines show you the result. When you search (or ask) something to a human, most of the times he may inquiry about your question several times and after confirming about your question, he will say you a result. If you search with the keyword “apple store” search engine will not understand what you are searching for (fruit store or mac store), as well as this search query also confuse a human also, that’s why he ask you 2^nd time. Personally I always wanted the search engines with discussing feature. And the discussion will not with some predefined questions. A dynamic discussion between user and search engine such like you are talking with an intelligent agent.

2. New approach; Real intelligence:

 

Soul algorithm introducing new concept inspired by natural learning system. For example if I say “an apple”, an impression will appear on your mind about an apple. This “impression” is the main concept of soul algorithm. And now if I ask you some valid questions about apple, you will give me valid answer and also you will reject any invalid statement about apple. Like if I say “I saw the apple was reading newspaper”. Though grammatically correct, you will reject the sentence, because when you read the sentence, some impressions on “apple”, “reading”, “newspaper” will appear on your mind and these impressions will combindly make a rule and that rule will reject the sentence.

How impressions are created about an object in our mind, how rules are created from impressions and make relation with each other, how a decision comes from an impressions and how new rule or information is added with an impressions, these issues will be handled by soul algorithm.

Unlike others, soul algorithm is not only doing pattern matching, calculating probability or weight adjusting for learning or decision making. It works on “impressions”. Whatever observed or input to this algorithm, it makes it an impression or update previous likely impressions properties. Soul algorithm understand itself and can feels it’s existence, and also it can make difference between unseen object properties like; good, bad, less, more etc. It observes any event like a living being and learn also as well as. It learns by its own system. Learning process is independent; language is not an issue for learning. If soul algorithm can recognize red color, it will recognize the red color with its own code, which we are saying “impression” about red. So whatever the color name in different languages, when an event comes with red color, soul algorithm will recognize the color by its unique impression code.

3. Expected intelligence level from soul algorithm:

 

Case 1:

Human: I see fire on my table! What should I do?

Soul algorithm: Do you have water, sand or extinguisher?

Human: No

Soul algorithm: Do you have anything liquid nearby?

Human: Yes, I have a packet of juice and some patrol on car tank.

Soul algorithm: Throw the juice on the fire.

Case 2:

Read the following article

“Any fruit juice can be a good base flavor for water, but tart juices, like cranberry, pomegranate, grape, and apple, are especially delicious. Go for juices that are all natural, with no added sugars. And remember: Fruits and their juices don't just taste good — they contain vitamins and antioxidants that can benefit your health too.”

Suppose that soul algorithm found this article for the first time. The algorithm now make new impressions or update previous impressions from this article and then will be enable to answer following questions or can give decisions on following situations, which traditional natural language processing technique will not be able.

Human: I have diabetics; I want some sweet tasting drinks?

Soul algorithm: Have some fruit juice

 

 

Human: How to change the flavor of water?

Soul algorithm: Why you want to change the flavor of water?

Human: For drinking

Soul algorithm: Will you drink yourself?

Human: Yes

Soul algorithm: Fruit juice can change the flavor of water.

4. The Concepts behind “Real Intelligence”.

 

When we say human intelligence, here appear two factors for dealing intelligence- these are

1.   Five basic senses

2.   Predefined or natural knowledge or instinct 

Let's see how these two factors are playing in the “real intelligence” system -

4.1. Five basic senses:

When any object exists or any object’s activities happen around us, our five basic senses (1. Sight, 2. Taste, 3. Touch, 4. Sound, 5. Smell) observe that object and the object’s activity. While observing any object or activity our senses set different sense values according to sense type; for example let us consider a new born baby who has no knowledge about this world, and now imagine in front the baby, there is an apple on a plate.

The new born baby! who has no intelligence, he can only see something or hold something around him, and there is an apple on a plate in front of the baby. And obviously the baby does not know anything about “Apple”.

Let’s observe the baby’s activities:

1.   The baby crawling to the apple,

2.   See the apple

3.   Hold the apple

4.   Take up the apple from the plate

5.   Smell the apple

6.   Bite the apple

7.   Get a sweet taste of the apple

 

Now the five basic senses assign some values for the “Apple” (for simplicity we are only considering the very basic observations and avoiding the plate, later we will consider the relation between the apple and the plate) in following ways

 

Sight => color=radish; shape=round

Taste => sweet

Touch => hard

Sound => no sound

Smell => sweet

Now the baby acquires some knowledge about that apple and in future when he will see another object with similar sense properties, he will treat that as apple. And reversely we also find another scenario that - after assigning the sense values the baby himself automatically take a name which can help him to uniquely identify apple from other objects. As the boy is very little and has no knowledge about the world and even he cannot pronounce perfectly so we found that after recognizing an apple, he is calling it as “apalam”. Now if someone say “apalam”, the baby get an impression of a round, radish and sweet taste object.

After time passed when the baby grows up and now can pronounce correctly and when he becomes hungry, he can feel it, someone learns him that “it is called apple which you were addressing as “apalam” and it is a fruit, when you get hungry you can eat it”. The kid has learned now the object apple with the same sense values.

Now let’s consider the kid found another object with the following sense properties

Sight => color=orange; shape=round
Taste => sweet
Touch => hard
Sound => no sound
Smell => sweet

He found that the 2nd object has almost same sense properties as 1st object except the “color”. So he can easily come to a decision that it is also a fruit when he becomes hungry he can eat it. And later he comes to know that the 2nd object was an orange.

(Here for simplicity complex & detail sense properties have been avoided, only for pointing that an intelligent system can be developed using our basic sense values.)

Now read the word “Apple”. After reading this word, did you notice that an impression raises in your mind like that “A radish color round sweet fruit”. Right? Now if I tell you “Orange” another impression will be recalled like “A orange color round sweet fruit”. These impressions are actually a set of sense properties, those were made by you when you see any objects, then you make group according to sense values. For example; you have already made a group named “Fruit” because you got the similarities in the sense values of the group members.

And also if you observed any rule for any object of the group, you will set the same rule for all members of the group. For example if I say that “Apple juice can be a good base flavor for water”. You will also learn a new rule that “Orange juice also can be a good base flavor for water” because you put the apple and orange in the same group.

In this way you are developing your knowledge base by using existing knowledge and without any target like other AI algorithms.

Now again let's come to the kid who can recognize fruits only. When he hears about an unknown object; he will want to know the sense properties of that unknown object. In this case he will do query like following, for example if say him “Car”, he will want to know

What is the color of car?

Does it taste sweet?

How it will feel after touching?

Has it a good smell?

Above queries are related to sense properties of a fruit. Because the kid’s knowledgebase has developed from recognizing fruits. When the kid will observe another type of object with his five basic senses, like “car” and also activities of the object, there will add a new impression in the knowledge. And accordingly similar objects will create different groups according to their sense properties and activities. 

4.2. Predefined or natural knowledge or instinct.

Let's go back to previous section, a newborn baby’s experience after seeing an apple. Let’s review the activities again - a newborn baby sees an apple and following events have occurred-

-The baby crawling to the apple,
-See the apple
-Hold the apple
-Take up the apple from the plate
-Smell the apple
-Bite the apple
-Get a sweet taste of the apple

 

Let’s consider that the apple and the plate have placed at right side of the baby. So after seeing the apple the baby wants to move to right side of him. After reaching at the apple he has to hold and put the apple from plate and take it to mouth to taste it. When he gets a good smell and sweet flavor, he will decide to eat the apple. But if the baby got any object with bad smell and flavor, he will reject it.

Did you notice that, here some activities performed by the baby those were not trained him. For example the apple and plate was at the right side of the baby, so he moved to right, but no one learnt him such lesson like that “if any object situated to right side then you have to move yourself to right side”, same way he put up the apple from plate to bite it, here no need to learn the baby that “you have to hold the apple to put it up”. After putting up the apple from the plate the baby will automatically discover the relation between the apple and the plate, that is “on”, that means “apple on the plate”

Here we found that the baby already has the knowledge of directions (right, left, up, down), for that he knows that he has to move right if anything need to hold that is on right side. He can take some decisions without learning anything, like if he got any bad smell or flavor from something, he will not eat it. He can understand relation between objects like an apple on the plate or plate on the table. Predefined knowledge base is playing the role to understand these kinds of relations or taking decisions just based on senses value (Like bad smell objects cannot be eaten or moving right to get something on right side). Some other example of predefined knowledge base are like counting knowledge - If anyone does not have the knowledge of number system, though the can count, suppose there were two apples on the plate and one apple has taken by someone in front of the baby, though the baby does not have the number sense, he will understand one apple is missing between two.

Now we will see how the soul algorithm use the above concepts to learning anything, taking decision, acquiring new knowledge using existing knowledge, feeling something (sorrow or joy, justice or injustice), and finally we will explain why it is named soul algorithm, and also why it is not only a real intelligent algorithm, but also it contains a soul like living being. That's why it can learn itself; no need to set it a target to learn about that.

5. Impression Network:

 

Let us think another simple scenario that the discussed kid saw someone took apple from trees and he also observed that trees are planted on earth… From this scenario the kid will get some new impressions like “trees”, “earth”, “leaves of trees”, “apple flower” etc.

Now the kid has a predefined knowledge base, this knowledge base can only help him to make relation / decision from the impressions he gets by his basic five senses. So whenever he experiences an impression he will put it in an orderly network; a network of impressions. See the following chain

 

trees - PLANTED ON - earth

   |

GROWS ----------- apple leaves

   |                                     

flowers                                

   |                                       

GROWS                             

   |                                        

apple ----------- HAS ----------- apple leaves

 

On the above network small case words are impressions and the capital with bold words are predefined KB values. So impressions making a relational network with predefined knowledge base. So whenever we experienced a new impression, we put the new impression in the existing network or start a new network and finally we are in a continuous process of building a network.

From any branch of the network we can extract information, logic, and decision and generate new knowledge by combining different types of impression. If we can parse the impression network properly by natural language processing, we can get valid statement from the network.

Apple cannot fly

The above statement can be found if the “apple” and “bird” exists on the network.

This concept will be clearer if we categorize the impressions around us, like “swim”, “eat”, “fly”... these are activity type impressions. We can understand any object is flying or swimming only by seeing or only by touching we can identify any moving object.

That means always we do not need all our five basic senses to get any impression, like only using our sight sense and touch sense we can get any activity type impression and explain the impression by relation with other types of impression through predefined KB values. For example, we can explain the impression of “swimming” following way

Swim = move in water by movements of arms and legs or tail

Here underlined words are impressions and bold words are predefined KB values

6. Making Group:

 

During learning or taking any decision, we first learn or observe any object’s rule than primarily we come to know that this rule may be applicable for same kinds of objects. In this session we are talking about impressions, so if two impressions have the almost similar sense values, we will categorize them into same group. So we can put apple and orange in the same group that is “fruit group”, now if we learn that “Orange juice can be a good base flavor for water”, we can come to the following decision-

Any fruit juice can be a good base flavor for water.

7. Impression Lifecycle:

 

Objects around us have life cycle. When we sense any object, we get an impression of the object for a particular stage of its life cycle. For example when we see an apple, we only see the apple. But the previous stage of the apple was apple flower and finally it will be eaten by someone or get rotten. So objects are changing their states and formats. Like our five basic senses there are some basic factors are responsible for changing objects. They are-

1.   Force / Pressure

2.   Temperature

3.   Light 

...etc

And these basic factors are always applied along with respect to time. Like after applying particular unit of force for certain period of time an apple can be turned to juice by mixing particular unit amount of water.

Actually any object is always affecting by those basic factors. Events around us are occurring for applying these factors. Like if water is an impression, any event occurring by water is the output of applying of those basic factors. Like trees are shaking for wind (pressure), Rivers are following for water pressure (gravity). We are feeling uneasy for weather changing (temperature effect) and so on… 

8. Soul Algorithm to NLP:

8. Soul Algorithm to NLP:

 

As discussed earlier, from any branch of the impression network we can extract information, logic, and decision and generate new knowledge by combining different types of impressions. So now consider the sentence “Apple juice can be a good base flavor for water”. This statement should not be stored in the impression network like such a nice well grammatical format. Simply consider that our network has the impression of “apple”, “juice”, and “water” and these impressions are staying in impression lifecycle in a database table format

impression_table        impression_lifecycle_table

id                                id

type                            imp_ref

sight                           time

taste                           applied_factor

touch                          sight

sound                          taste

smell                           touch

identity_code              sound

                                   smell

Now the algorithm found a new object “Apple” and put its initial sense value to impression_table, at the same time a lifecycle will be started by taking ref id of the “Apple” in the impression_table. So in the time and applied_factor field of impression_table, we found some unit value like 5 unit in “time” field and 10 unit pressure in “applied_factor” field. And also we will get new affected values for “sight”, “touch”, “sound”, “smell”. Simple saying if any factors (like time, pressure, gravity, heat ...etc) applied on an impression we will get some changes and these changes are continuous process. We see a green apple ripe and turn into radish color after passing certain period and effect of certain weather affect (temperature). For example if we consider the apple ripening process, we will see with the passing of time an apple grew bigger (size change in sight affect) and radish (color change in sight affect), so if we want to keep the event in the above database tables, we have to put different sense values for different unit of time values for the same impression reference. After that we can extract any information like “An apple needs 10 days to ripe” by applying NLP on the database. That means we are storing data or building the impression network according to soul algorithm and extracting or use information by using NLP.

9. Concept of Soul:

 

Till now we have discussed about an algorithm and its input and output and some data processing technique. But now I will explain the idea of the “soul” part of this algorithm, and that’s why it will be real intelligence algorithm. The main disadvantages of traditional artificial intelligence algorithm are, these do not know what to learn, why to learn and most importantly machines are unable to handle emotional matters. Traditional artificial intelligence is just a machine. For example - suppose we are running a NLP system which can communicate both way - that means it can take input from users and analysis the input and deliver a reply or info (Like clever bots). Now if you tell this NLP system “Five soldiers died”. Definitely the system will reply “Alas! Very sad!”. Now you tell is the reply is ok? Or the analysis of the system is ok? For the traditional AI system, it is OK. Because these algorithms works only based on their comparison capacity, in the above case the system will analysis the word “died” and as this is the word related with pathetic issues, so the system answered so. But the answer is not correct for all cases.

Later if I add that the “they were our enemy soldiers”. In that case the answer should be “Great!”, well done!!” or something like that. This is the one of the main feature of soul algorithm. For any event related impression changing effect it will assign a soul value, if the value is higher than the events or impression is going on safe line. Let's explain-

Suppose we are handling this issue by a DB table named soul_table

Soul_table

Id
ref
owner
decision_point
target

As discussed earlier, if 10 days (time) applied on an apple then it becomes ripe, its smell, taste, looking will be changed and ok for eating for you and objects of your group (all human being). So in this case the values of the soul_table will be like

Id = unique id
ref = event ref (from empression_lifecycle table)
owner = for whom this event is occurring
decision_point = initial value ‘0’, if the event is positive it will go higher and lower for else
target = greater value from current value. It will use to find better option.

For example if we set the value of “curiosity” impression “10” for any event then the target value will be “11”. When any applied action go positive and the curiosity value will set to “11”, at that stage target of that event will be “12”, In that way the soul’s curiosity target will always increasing and always find a better solution and enhance its Knowledge Base.

In the above case if we get more sweet and good apple for “11” days’ time applied factor, it will increase the value of decision_point higher.

Now consider another case; if the applied_factor = 30 days and we get a rotten apple’s properties like bad smell, bad taste in the sense values of impression_lifecycle table, we will get negative value at the decision_point, and this value will gradually down after getting started of negative events like starting bad smell, starting bad taste to eat ….etc

At the same time let’s consider another case - that some insects like rotten fruits, so at the decision_point, for rotten fruits values will be decreased when owner = human group and values will be increased if the owner = insect group.

Now time to initiate a soul. Consider the case of a baby boning. After a baby has boned, he or she got an identity. Here identity is not meaning the baby’s name. That means every baby is unique like DNA profile. During growing the baby, every moments he grouping the impressions he got, like my mother, my father, my family, my safety, my family’s safety …. Etc.

Recall the making group section of soul algorithm, and also consider the last section where we discussed about decision_point.

OK. Now create a soul and it’s unique id is suppose 28022017124714. Let’s initiate it. While initializing the soul, simply we have set the following issues. In the soul_table the owner value will be “28022017124714”. To avoid complexity, simply the algorithm should be as following –

10. The Algorithm

 

Before this section one question must be knocked you that is - soul algorithm will work on five basic senses, but current latest technology not yet so smart even unable to provide us any sensor which can sense perfectly as human do; we could not invent smell and taste sensors. We are using sight, touch, and sound sensors but these are not giving so accurate data to make difference between mostly similar things or events. So the question is-

 In this situation how soul algorithm will work?

The solution is “Alternation”. Suppose, someone describing you about a spice, which you did not taste before. If you described perfectly, you can learn about the spice and you can use the unknown spice appropriately to make food. So here you have learnt by hearing and without using your sight (eye) sense and taste (tongue) sense. As we do not have the sensors so we have to develop the impression network with all possible instance of basic senses data.  Simply in the impression network database we will input what we see, but important thing is maintaining standard and pattern of inputting method. We will try to assign almost similar value for any impression or event, like hexadecimal code for colors.

10.1. Algorithm Prerequisite

1. create unique soul [with unique ID using date time]
2. create & initiate senses [which sense for what input during input]
3. develop & define predefined KB [relation & rule like; IS, HAS, MOVE, GOOD, BAD, GRAVITY]

 

10.1.2. Soul Algorithm

01. ask sense value for the impressions. (assume that there is no stored impressions). If match with any existing impressions, then update impression network

02. establish relation with other impression by predefined KB.

03. update impression network [place the impressions maintaining parent & child relation].

04. register the impression to impression_lifecycle process.

05. publish & register rule / result for every single applied factor.

06. store rule and info, update result in soul_value.

07. register / update group_network.

08. apply stored rules group wise on all available impressions.

09. filter the exceptions and store exceptions with corresponding rules.

10. permutation & combination among rules & impression for new and better Knowledge and update target of soul value for existing knowledge

11. maintain parent and child while storing rule or info.

11. Soul Algorithm Description:

 

Algorithm Prerequisite setup:

11.1       Creation of a Soul

New soul created, assigned id is 20170314151015bdctg.

11.2       Initiating senses

20170314151015bdctg soul has five senses as following

20170314151015bdctg.Sight  

=> Can take input for Shape (Value: round, rectangular, triangular etc.)

=> Can take input for Color (Value: red, green, blue etc.) 

20170314151015bdctg.Taste

=> Can take input for Taste (Value: sweet, sour etc.)

20170314151015bdctg.Touch

=> Can take input for Touch (Value: hard, soft etc.)

20170314151015bdctg.Sound

=> Can take input for Sound (Value: loud, quiet etc.)

20170314151015bdctg.Smell

=> Can take input for Smell (Value: sweet, bad, chemical etc.)

Now we have to setup scale for each value for each senses, for example let us consider about color; the hexadecimal color value are different for red -> more deep red – more dark red. Same way our value of unit scale will be different for like sweet-> more sweet, loud -> more loudly

11.3     Predefined KB development

As discussed earlier here we set up predefined or natural knowledge or instinct, here we will define instinct type and value like as following

Define Identity

Identify self (20170314151015bdctg) or any impression or event by unique vale

……… etc 

11.3.1 Define Relations

20170314151015bdctg.PKB.Relation.Direction

=> Can take input for Direction (Value: right, north, up etc.)

20170314151015bdctg.PKB.Relation.Connection

=> Can take input for Relation (Value: have, is, of etc.)

20170314151015bdctg.PKB.Relation.Owner

=> Can take input for Time (Value: parent, child etc.)

……… etc

11.3.2 Define Unit

20170314151015bdctg.PKB.Unit.time
=> Use to process time (Value: min, sec etc.)

20170314151015bdctg.PKB.Unit.length
=> Use to process length(Value: M, CM etc.)

……… etc

11.3.3 Define Rules

20170314151015bdctg.PKB.Rules.Time
=> Can process time (Format: initial_state, applied_factor, affected_factor, result, exception)

20170314151015bdctg.PKB.Rules.count
=> Can process count(Format: initial_state, applied_factor, affected_factor result, exception)

……… etc

11.3.4 Define Felling

20170314151015bdctg.PKB.Felling.Hungry
=> Can process Hungry felling (Format: flag, reason, scale)

20170314151015bdctg.PKB.Felling.Happy
=> Can process Happy felling(Format: flag, reason, scale)

……… etc

11.3.5 Initiate Soul value

self _id= 20170314151015bdctg
good_value = 50
good_value_target = ++
happy_value = 50
happy_value_target = ++
bad_value = 50
good_value_target = --
hungry_value = 50
hungry_value_target = --

……… etc

In this section we initiate the nature of soul, if we consider usual thinking, a human always find his own good and try to increase his good or positive things and also with this try to avoid or decrease negative events. For implementing these cases we have to initiate the soul values. 

11.4. Soul Algorithm: Illustrating

A simple soul has been created and it will start learning.

To avoid complexity we will ignore some issues mostly related with predefined KB like gravity, space etc.

As discussed earlier as we do not have any smart sensor so when this algorithm learns something, it will use our eyes or ears or nose. That means when the algorithm ask input, we will input data according to its format.

Let’s see a case; in this case we are going to input data about “apple” considering that the apple is unknown. So according to algorithm

Step 1: For any new impression, the algorithm will ask the values for each basic senses like for sight sense it is round and color is green (considering initial stage of apple), it has good smell, it tastes sweet, it is kinds of soft to touch, it is not generating sound…...etc. If basic senses matched with any existing impression like suppose the impression "water" is already exists in the network, then we may have to create any new branch of the network like "source" or "parent" of the impression. For example we can get water from river and also from rain too.

Step 2: Here we link the impression “apple” with other impressions like

apple->grows-> from->trees. Considering that the impression “tree” is already stored on the database and the definition of  “grows” and “from” are already defined on predefined KB.

Step 3: In this step we will update the impression network with the new impression “apple”. As we observed apple is growing from trees so the impression “apple” is the child impression of parent impression “tree”. So we should create a new branch of existing impression network from the existing member of the network “tree” and put the new impression “apple” other side of the branch like following

existing impression network ->tree->grows->apple

Step 4: We got a new member in the impression network, now we will register the new impression “apple” in the impression lifecycle. In the impression lifecycle different natural factor like light, time, temperature, will be applied unit wise and the result after applied factor will be updated on the impression networks on associated impressions and associated branch. For example after certain period of time a green apple becomes red and sweeter. Whenever we got different result from initial result we will update the impression network. So we got a result after applying some unit of time and temperature on the impression “apple”. Now we should update associated impression and branch. For the discussing example the associated branch should be update most likely following ways-

existing impression network ->tree->grows->apple->20 days->red

And more sub branches will be generated from the impression “apple” like

apple->20 days->bigger

apple->20 days->sweeter

apple->20 days->softer

Now if we get another impression let suppose “juice”, then we will update the existing branch of existing network in following way

existing impression network -> tree->grow->apple->pressure->juice

Here “pressure” is predefined and after registering the new impression juice in the impression lifecycle and after applying natural factors on “juice”, more sub branches will be generated from the impression “juice”.

Step 5: Before starting this step let's take a look at human learning process. Whenever we experience any new events around us, we memorize the event category wise. Let's explain; suppose we are observing orange juice making process for the first time, in this event there are two parts as following

Info Part: color of orange juice is “orange” (ignoring others)

Rule part: if pressure is applied on fruits, it turns into juice. (ignoring others)

So from impression lifecycle, we extract information and rule for every applied factor and store these for using into next steps.

Step 6: At this step we will work in the core of the algorithm, as again let's see another side of human learning process. In every moment we are experiencing new and repeated events, some events are liked by us or disliked or some events are useful and some are harmful. We always are accepting the positive events and want more positive and better for anything. This is our soul. Our soul likes good, want to be stay good and expect to be better for any case. Our soul is curious and when its curiosity satisfied it becomes more curious. These cases like fillings good, bad, like, love, hate, couriers are treated as soul value in this algorithm. In the impression lifecycle for every result of applied factor we will update the corresponding soul value. For example for a ripe apple tasting we will get good filling and will experience bad filling for rotten apple. So in the impression lifecycle when applied factor on a green apple is 20 days and result is a ripe apple and it's “taste” value is good, we will update the soul value of this entry in the positive direction at the same time we will also increase the target value so that the algorithm always seeks a better result for any case. Same way we will do opposite for a rotten apple decrease the target value.

Step 7: This an important step to increase knowledge using existing knowledge. For example if we know that “ Orange juice is good for health” then simply we can say “Apple juice is good for health” and it is a valid knowledge. We can generate this new knowledge because we know that the orange and apple both are fruits. So for generating knowledge from existing knowledge, we have to categories the impressions. In this case the impressions those have similar or close sense values and giving similar output after applying same factor in the impression lifecycle, these impressions will go under same group.

Step 8: We already categories all the impressions according to their nature and reaction against basic factors. Now whenever we get discover new rule or info with any impression, we will apply the rule to the all members of the corresponding impression group. For example the impression “water” is the member of “liquid” group. Simply we know that water can put out fire. so PRIMARILY we can apply this rule on the all members of liquid group ( but the rule is not always true for all cases; these exceptions will be handed on the next step). Now we found that juice is also a member of liquid group, so according to algorithm can we say - “juice can put out fire”?. Yes we can and it is a valid rule. 

Step 9: As we discussing earlier if we apply any impression’s rule on the all members of that's group, there will be some errors for exception. For example petrol or diesel can be primarily registered under liquid group, but these are not totally appropriate for putting out fires like other liquids. So for every generated rule there should be “exception” field which will be updated time to time.

Step 10: Let's go back at the beginning of the article to the new born baby. When a baby learn something new then he applies his new learning to another new things he found, for example we observed that a baby try to eat everything he found around him. He accept it if that thing was suitable to eat and reject otherwise. Same thing happening with other animals also. Living being are learning by permutation and combination of their existing knowledge. Let's explain- consider the following info and rule

 

info: fruits are food

rule: something can be crashed by applying force

in nature we see that when animals or birds found unknown hard fruit, they crash it and try to eat and next time when they found same fruit, they will do the same thing. They have learnt this new knowledge by combining above info and rule.

Let's see another example. We have the impression “water” in our database and also the rule “water can put out fire”. And we have also some other impressions like “apple”, “orange”, “tree”...... etc.

Now add a new impression “carbon dioxide”

Now combining the impression and rule

apple can put out fire- no

orange can put out fire- no

carbon dioxide can put out fire- yes

Now come to the most important point! We are getting new knowledge or info or rule by combining among the rules, activities and impressions. For every result of the combinations or any natural events (like rain drops make sound), something inside us assign a value for that event or knowledge or info. That value is treating here as soul value. While tasting ripe apple we feel good, so value for that combination or applied factor on any impression will go upward and for rotten apple it will go down. At the same time a target value will be set for each cases higher than current value so that the algorithm always search and get the more efficiency. Just we have to defined the rules like positive events or emotion value will have a target of  increasing and vise versa for negative events  in predefined KB section.

Step 11: Actually this step should be applicable whenever we store any impression or rule in the knowledgebase network, For example we will maintain the branches and sub branches of impression and rule network wings wise, like “ yes wing” and “no wing”. For the above case rule 1 and 2 will go under “no wing” and rule 3 will go under “yes wing” and also always put the impressions maintaining parent and child like

tree->grows->apple

water->put out->fire

this structure will help to generate valid sentence like if we bind the branches not to go reverse then we can say

tree grows apple is valid

and in opposite direction

apple grows tree is invalid (ignore exception)

Also

water put out fire is valid

and in opposite direction

fire put out water is invalid (ignore exception)