Designing a computational Approach to understand Smell & Taste

Abstract

The of topic this dissertation explores different areas of research in the fields of software engineering, computer science, biology and bio-chemistry. The purpose of this research will contribute to creating the design of a smell and taste system. The researcher implemented an exploratory method of research to understand the origins of the five senses. Computational counterparts of the five senses will be researched, which allows the researcher to triangulate data as there are multiple similarities. The researcher has also researched smell & taste and its closest computer replicas. Knowledge on artificial intelligence algorithms allowed the researcher to understand how a computer processes information to the logical standard of a human being.

Smell and taste can translate thousands of different molecules, into one specific taste/smell. Like the three replicated senses (Vision, Hearing and touching) transduction Is a key process, as it translates molecules into electrical signals for the brain to understand. Classification of smell and taste can be defined from a combination of many aspects, such as:

• The five tastes – Sweet, Bitter, Sour, Salty, Umami • Chemical Structure • Texture • Temperature • Culture

The design of the researchers’ system will be able to distinguish between different types of liquid-food and drinks using its chemical structure (Soup, coke, juice etc.). the machine will use a supervised learning approach to ensure accuracy Is achieved when processing results. Algorithms contributed towards designing a taste and smell piece of software by distinguish between different types of liquid. As this paper is based on secondary research, dataset examples from other researchers will be explored.

Finally, like many other computational counterparts the design has its advantages and disadvantages over the biological counterpart, enabling customization/add-ons of the future design. Overall Researching how the sensory systems work from a human and computational aspect (Hardware & Software) will be concluded in this dissertation. A test case and interpreted design will conclude the researchers’ dissertation.