Introduction
Artificial Intelligence (AI) is currently traversing through its natal stage. The implementation of AI shall lead towards analyzing the cancers and tumors better than human inference, it is also leading towards solving various criminal cases while providing entire rule base of the legal framework and it can view and capture various images and photos and can classify the person as in social media platforms and various real-time image capturing devices as well for autonomous driving cars etc.
It is also enabling to initiate a chat with the user via function-specific chatbots using NLP to perform various cumbersome jobs. Soft computing has initiated the process of integrating various mathematical processes and technologies used for analyzing data. These baby steps in silos are towards the creation of the self-reliance of a holistic system. We have already started working with various types of datasets lying under different categories. Application of fuzzy logic, genetic algorithms, neural networks, and hybrid systems are incorporated in various technical and economical perspectives worldwide. Artificial General Intelligence (AGI) is explaining the intelligence of machines required to act like a human during various scenarios.
But, still, we have to go leaps ahead in the direction of achieving actual intelligence which is inbuilt in any living thing. Most of the data generated at any event goes in vain at every moment, which shall give altogether a new perspective while taking decisions in our life. Decision making using natural intelligence is based on data captured as facts of that moment. Similarly, it needs all the perspectives prevailing at that moment in that scenario to be captured for decision making using artificial intelligence as well. The gut-brain model comprises hierarchical semantic databases that shall capture the dimensional data and use it in a realistic manner as expected by us.
Working of Gut-Brain model
The conceptual gut-brain model proposes a full-duplex communication and integration with the holistic repository of cloud and big databases, extracting the linkages amongst cognitive centers of the repositories and linking emotional actuators using various perspectives with every profile under consideration. Currently, we are surfing an era of programming intelligently rather than the real perspective that leads towards intelligence incorporated through dynamic programming and re-routing while analyzing real-time perspectives. Just, like Sophia, the robot is not intelligent using AI, although it is intelligently programmed to answer the question asked.
The figure-1 shows Gut -brain model for artificial intelligence in which VirMon (Virtual monitor) is the most important function towards communication between neural networks adjoining towards functional disparity while handling data streams. These data streams are classified at the level of ARC which is Augmented Reality Corresponder, it constitutes of positive data intake using NPY and AgRP where NPY is leading towards Non Physical Yield of data like stress, moods, dullness, anxiety, and depression etc. AgRP Augmented Reality Protocols leading towards the reality based data streams. Negative data intake shall be taken care by POMC which is Portable Object Mediation Connection provider. The data streams from ARC leads towards PVN which are Pure Value Neurons dealing with purity of data induced from ARC. These pure value neurons shall get registered with hierarchical semantic centre via hedonic mediators which are eagerly waiting towards pure data for further relations. VirMon has another set of data streams induced via TAE known as Target Activity Enhancer, which comprises of PYY deals with purity of your yield to learn and GLP-1 which generates learning partition for the data induced via L-Cell which is learning cell using non-hierarchical data.
Benefits of Negative data and use of the Gut-Brain Model:
Artificial intelligence is moving with the help of a classification idea. The idea may be negative and positive data. Example Smiling face may be producing positive data and, frowning face may be negative data. Artificial intelligence mostly use the positive data for learning, but there’s much more use of negative data for future accurate predictions. It shall be explained with the help of examples only, which are happening in our lives almost on a daily basis. The following benefits can show the negative data relation in predication.
1. Treatment of patients with cancer or any critical diseases:
The positive and negative data helps doctor for making predictive model for the diagnosis and treatment of diseases. The artificial intelligence is used to collect patient medical , ethical, legal and social data, these may be positive and negative. But this data help doctors to make decisions for the diagnosis of the patients.
2. Web search engine result refining:
Search engine like google, wiki, duckduckgo, bing etc will work well if they have both the data records. These search engines are using both types of data, but in unequal proportion. If the data generated is used in the same proportion that will impact a lot for decision making.
3. Spam Email and Filtering:
The Gmail and other mail server using both (positive and negative ) data with almost the same ratio for spam detection. This can improve the system security in day to day life.
4. Fraud Detection Online:
With the help of the negative data, we are able to predict the fraud, happened in credit, debit card and in the banking sector. The fraud detection is one of the most important applications in today’s world, and to remove with only be possible with the help of all the data available with us, either in negative and positive forms.
5. Making Business Decision:
The industry is dealing with a huge amount of the data, the data may be in any form (positive and negative). The business will run more efficiently if we know the all factors in well manner that says the positive and negative impact of decision. It may help in better planning for financial, marketing and sales.
The above all benefits and more like voting machine learning, Sports match prediction, government decision taking etc. may use the data in equal amount. The Gut -brain model can help for these types of decision. They use the real time data, either its positive and negative. That help to make a decision much more accurate and visible to end users.
Challenges in Gut- Brain Model:
1. To get the correct data:
The data that we find is always conflicting. So we need to categorize the data with their functions.
2. How to make sense for any data:
The major challenge is to associate the data with correct sense and group. Due to huge and variable nature of data, it is difficult to map the data in correct sense.
3. Technology challenge:
The technology is a big challenge because due to their complexity and expensiveness. The developers have always been consider this thing into mind when developing the algorithm regarding same.
4. Research Challenge:
you have to remember each and every factor doing research. The present scenario is towards the positive data, less amount of research on negative data. So collection of data regarding same is one of the biggest challenges.
5. Data Regulation challenge:
One of the important challenges is data regulation, for Gut- brain model working, we need a real time data for doing analytics. The real time regulation of data become much more challenging in predicting the decision.
Conclusion
According to the gut -brain based model each and every data is important either it’s negative or positive. The gut-brain model can be used for communicating between the devices to share the data. The system works with the real time data, which may be used for accurate prediction. The learning with this model will help for many applications like medical for patient treatment, web mining for finding accurate results, removal of fraud in communication and many more. The model can improve the scope of the artificial intelligence for prediction in real time scenario. There is some technology and research challenges exist to do the same, but we are able to face and provide the real data analytics for better future prediction.
References
1. https://www.nextbigfuture.com/2018/12/will-mimicking-the-nervous-system-advance-artificial-intelligence.html
2. https://www.forbes.com/sites/cognitiveworld/2019/06/10/how-far-are-we-from-achieving-artificial-general-intelligence/#4a8368cd6dc4
3. https://en.wikipedia.org/wiki/Artificial_general_intelligence
4. https://whatsnext.nuance.com/healthcare/is-artificial-intelligence-healthcares-new-nervous-system/
5. https://www.humanconnectome.org/
6. http://www.humanconnectomeproject.org/
7. https://www.nextbigfuture.com/2018/12/will-mimicking-the-nervous-system-advance-artificial-intelligence.html (Biologic Intelligence)
About the Authors:
Dr. Durgansh Sharma, currently working as Associate Professor, Computer Science and Engineering, University of Petroleum and energy studies, Dehradun. He has conducted various workshops, conferences MDP and FDP. He Guides various students for research and project work. He is an authored 25 research papers, 1 consultancy project. His academic interest includes Image Processing, Machine Vision, Robotic Process Automation, IoT systems, Healthcare, Energy Distribution Automation. He can be reached at [email protected]
Dr. Sunil Gupta, currently working as Professor, Computer Science and Engineering, University of Petroleum and energy studies, Dehradun. He has conducted various workshops, conferences and FDP. He Guides various students for research and project work. He is an authored 65 research papers, Four Patent and two books, namely Cryptography and Network Security and Wireless Sensor Networks. His academic interest includes Security, Cloud Computing, Big Data, Sensor and Wireless Networks, Healthcare. He can be reached at [email protected].
