DESC:Technical field
The present invention pertains to the decision by the artificial Intelligence (AI). More particularly, it relates to the decision making by AI based on the parameter of emotion like trust and prevailing information like advantages of a product / object or object in context as observed in human beings.

Background of the invention
Artificial intelligence is the field of computer science concerned with creating a computer or other machine which can perform activities that are normally thought to require intelligence.
One subfield in this area relates to creating a computer which can mimic human behavior, i.e., so that the computer, or a character displayed by the computer, appears to display human traits.
A substantial amount of effort has been made in this latter area, i.e., to provide a computer character which appears to display human traits. Unfortunately, however, the efforts to date have generally proven unsatisfactory for a number of reasons. Some of these are, the artificial intelligence program customized for each character, which is a costly and time-consuming process; customization of AI for specific application program; etc.
It's quite difficult to define precisely the word decision, but each of have already experienced the concept. Every human being, right or wrong, thinks that he has made a choice between different alternatives. Whether he exercises his free will or bends to some kind of causal necessity, is another (philosophical) question. The intuitive notion of human free will in choosing between various alternatives. On the other hand, it is important to specify what we mean when using the expression "Artificial Intelligence" (AI). There are at least two different views about AI. The first one equates AI to the science of designing and building computer-based artifacts performing various human tasks. Adopting this view, curbs the philosophical discussions about the nature of intelligence and the feasibility of the AI project. This view of AI has relatively few links with decision to the extent that an artifact cannot properly be said to make a decision. The decision, if any, has of course previously been made by the designer of the system. In other words, the concept of "decision" is antinomic to the idea of program. When a task is programmed, the decision no longer exists since the actions are predetermined according to each possible situation that may occur. But even if an artifact does not make any decision, its designer has previously modelled a decision process embedded in the system. And this is a first question for us: how to model and program decision processes in the artifacts? The most natural answer to this question is that "it suffices" to observe how people make the decision in the task at hand and to reproduce the process into the machine. So, even if we adopt a view of AI not referring to "human intelligence", we have to deal with human reasoning.
The present invention overcomes the shortcomings of the prior art to introduce the human trait of decision-making and provides the AI system and the process to introduce the decision making ability into the AI.

Summary of the present invention
The present invention discloses the AI bots system with decision-making learning based on the trust parameter between the AI bot and the correspondent. The present invention provides the system which enables the chatbot/virtual assistant to take the decision for any information feed, query or question depending upon the intensity of the trust factor between the chatbot system and the correspondent.
The decision making system by an emotion like trust (being influenced by the known population) comprises of the neural network of neural signature wherein a unique neural signature is assigned to the intensity/weightage of the trust with the correspondent. The trust is build up through the interaction of the chatbot with the correspondent (human / bot) and the information provided to the chatbot by the correspondent. The trust parameter is compared to the neural signature available against the options available. This neural signatures (percentage for current understanding in context) are variable wherein they are capable of upgrading or downgrading depending upon the process of the trust (trust is an emotion which also has its own neural signature) building with the correspondent. This parameter is associated with the object which here, is a correspondent which can be another chatbot or a person/individual. The trust parameter measured in percentage is coded into the neural signature wherein the number of the percentage is coded into the neural signature. This neural signature coding for the intensity/weightage of the trust in percentage is then associated with the neural signature of the respective correspondent (object) in the neural network of the bot system.
The other disclosure of the present invention includes the decision making by the chatbots/virtual assistant wherein the decision is based on the advantages or the disadvantage of the object which, in here, is shared by a correspondent. The decision making in this scenario is based mainly on the information stored about the object. This information is also coded to form the neural signature which in return is associated with the respective object (correspondent).

Detailed description of the invention
The present invention pertains to the learning of the decision making by the chatbot based on the emotion of trust which is a parameter observed in human beings also. It also discloses the decision making by the chatbot based on the information of the object (correspondent/the object in context). The chatbot mentioned here can also be a virtual assistant wherein every chatbot/virtual assistant is initially provided with the supervised learning comprising of basic information for operation and functioning.
The present invention includes the chatbot/virtual assistant system that learns the decision making based on the trust as emotional parameter wherein there is building up of the trust between the bot and the correspondent during their interaction. The trust is measured using different neural signatures (for understanding in context can be referred as percentage). This intensity or the weightage of the neural signature of trust is coded to form a unique neural signature. The coding can be in any system like the hexadecimal system, binary system, octal system etc. The object (correspondent) is also coded into the unique neural signature in the neural signature network. Thus the neural signature for the trust intensity/weightage is associated with the neural signature of the respective correspondent with which the trust is build up. The trust builds up with reference to the information feed or the answers received by the owner or the creator of the particular chatbot/virtual assistant system. Suppose the correspondent (object) feeds some information or the answer to the chatbot system, the chatbot system first confirms the authenticity of the information or the answer fed to him with his owner/creator. If the information or the answer fed to him matches with that of the information/answer provided to him by the owner/creator for the same query, then there is the buildup of the trust with the correspondent. With each interaction there is upgradation in the intensity/weightage of the trust parameter through increase in the percentage number with the correspondent, if the authenticity of the information provided is confirmed by the owner/creator of the chatbot system. If there are information being shared about an object in context and the AI Bot is not aware of the same, the same is learned but not put in use for decision making till the acquired information is verified with people/correspondents with good trust values like the creator.
The building up of the trust parameter and its association to the respective correspondent is through coding into unique neural signatures which are linked up in the neural signature network. Suppose the correspondent for a particular chatbot system is a friend of the owner/creator of that chatbot. This friend and his details are stored into the chatbot system wherein it is coded to form the unique neural signature for that friend. The coding can be with any coding system like the hexadecimal system, the binary system etc. There is the trust development with this friend during the interaction of the chatbot system. This trust measured in the percentage is also coded with say hexadecimal system to form again the unique neural signature. With receiving of the information from this friend, it is first confirmed with the owner/creator by the chatbot system. If the information provided is authentic with reference to the owner/creator there is upgradation in the trust percentage which is the again stored with different unique neural signature and is linked to that friend in the neural network. This coded neural signature is retrieved from the memory at the next instant of interaction with this friend wherein the decision making is required.
In another embodiment of the present invention, the decision making task by the chatbot do not involve the trust parameter but is based on the information associated with the respective correspondent. This information is mainly determined to be advantageous or disadvantageous with reference to the context for decision to be made. This information is stored in the form of neural signature and is associated with the neural signature of the respective correspondent. When the decision is to be made for the context/subject which also matches with the information of the correspondent, the memory is triggered and the neural signature of that information is retrieved. The neural signature for the advantageous information is mainly retrieved which would help in decision making. In case of decision making wherein there is no neural signature formed for a particular query or the information, the bot resorts to decision making according to the law of the land or the rules of the game as perceived during the initial supervised learning.



Example 1
As an example for the decision making through building up of trust parameter; a neural signature for a FRIEND of the owner/creator of the chatbot system could be A13DC2AE which can have John and Rohan with child relationship. Both John and Rohan have their own unique neural signature e.g. 00013A09 and 0A1C3A8D respectively. There would be a parameter of trust which would have a weight age based on which decisions would be taken. More the weightage better trust relationship Bot enjoys with them. Suppose the trust parameter is 70% with John which is available in the unique neural signature say 00021A08. Now John feeds the information that “Sky is blue”; this information is coded to form a unique neural signature say 00024A05. This neural signature is stored in the memory file associated with John. Now this information is first confirmed with the owner/creator for its authenticity. If the information is confirmed by the owner/creator for its correctness, then the trust level for John providing the information in the first place is upgraded form 70% to say 71% which is stored with the neural signature of 00021A09. On the other hand, if the information provided is corrected by the owner/creator, then the previous neural signature for the information provided by the john is erased/ignored and the new neural signature is formed for the same information corrected by the owner/creator. Thus when there is a given input by John which has a better trust relationship, his shared knowledge would be applied and decision based on the input provided shall be applied.

Example 2
As an example for decision making based on the advantageous and disadvantageous information of the correspondent; say there is FRIEND of the owner with neural signature A134DE5C with two names John and Rohan. Both John and Rohan have their own unique neural signature e.g. 00013A09 and 0A1C3A8D respectively. Consider the information about Rohan wherein Rohan is a doctor. This information that “Rohan is a Doctor” is coded to neural signature say 001AED04. When there is a context or query related to medicine or health for which the decision is to me made. This neural signature of 001AED04 is triggered and accordingly the Rohan can be contacted by the bot system, since the information of Rohan being a doctor is already available with the bot in the advantage section of information about Rohan in discussion to context of medicine.
,CLAIMS:We Claim:
1. An AI bot system comprising of, a decision making function wherein the decision is about a human and/or an object.
2. A system as claimed in claim 1, wherein the decision making is based on the emotion of trust between the correspondent and the AI bot and information about a said human and/or object.
3. A system as claimed in claim 2, wherein the information about said human and/or object is fed by the owner of the AI bot.
4. A system as claimed in claim 1, wherein the said decision making enabled system also includes a virtual assistant.
5. A system as claimed in claim 1, wherein the said system includes supervised learning for the AI bot wherein the said learning includes the basic information for operation and functioning of the system.
6. A system as claimed in claim 1, wherein the trust build up between the AI bot and correspondent takes place during the interaction between the two.
7. A system as claimed in claim 1, wherein the trust is intensity or weightage of trust is measured in percentage wherein every trust intensity is coded to form the unique neural signature.
8. A system as claimed in claim 1, wherein the basic information pertaining to every human correspondent and an object is encoded to form a unique neural signature.
9. A system as claimed in claim 1, wherein the unique neural signature of a human/object and the unique neural signature of the trust intensity in percentage is linked to form the neural signature network.
10. A system as claimed in claim 1, wherein the trust build up is with respect to correct and authenticate answers provided by the correspondent to the AI bot.
11. A system as claimed in claim 1, wherein the authenticity of the correct answer by the correspondent is verified by the AI bot through the owner of the bot.
12. A system as claimed in claim 1, wherein there is up-gradation of trust intensity with every authentication by the owner for correctness of the information/data provided by the correspondent.