In recent years, we have seen
the hazardous development of the sheer volume of data. The quantity of books,
motion pictures, news, promotions, and specifically on-line data, is stunning.
The volume of things is impressively more than any individual can channel
through keeping in mind the end goal to locate the ones that they will like.
Individuals handle this data over-burden through their own exertion, the
exertion of others and some favorable luck. The development of recommendation
systems have grown parallel with the World Wide Web. In the initial phase,
these systems were based totally on demographic, content-based and
collaborative filtering 1. These systems are predicted to use the local and
personal information from the Internet of things for their computations in
future. Recommendation of information has turned into a vital research region
since the first paper on collaborative filtering which was published in the
early 1990s 2. The interests in this field has increased rapidly because of
the abundance of applications and the increase in users using these kind of applications,
for example the recommendation of books, products and other related items at
Amazon, also movies recommendation by Movie Lens, restaurant recommendation by
Yelp, and many more.
A recommendation system
gathers information on the inclinations of its clients for a set of items (for
example, food, movie, books, music, gadgets, travel destinations, etc.). These
information can be acquired by collecting the client’s rating (explicitly) or
by monitoring the client’s behavior on the recommendations (implicitly) 3 4.
The Recommendation System may utilize statistic highlights of clients (such as
age, gender, health, ethnic group, etc.), social information’s. There is also a
growing tend towards the utilization of data from the Internet (i.e. location,
real-time health signals, weather, time, temperature, etc.). Recommendation
system make utilization of various wellsprings of information’s for giving
clients with forecasts and suggestions of items. They attempt to adjust factors
like exactness, oddity, disparity and soundness in the recommendations.
Collaborative Filtering (CF) techniques play a critical part in the
recommendations, in spite of the fact that they are frequently utilized
alongside other filtering methods like content-based, knowledge-based or social
ones. The working of Collaborative Filtering is similar to how the human being
make their decisions on a daily basis. We tend to make our decisions not only
by our own experiences, but also on the basis of experiences and learnings from
a relatively large group of acquaintances 1.
As there has been a
significant increase in the number of people using the software applications
for searching for the things or places over the internet, the enterprise
offering these services are facing a challenge to dynamically maintain and
optimize the services provided by them 5. The goal is thus to design a
decision-making aid system to overcome the challenges faced during the dynamic
maintenance and optimization of the services provided by the industrial objects
during their lifecycle 1 5. Most of the automated systems which are
providing these kind of services do not give the means of intelligent
translation of data duplicating with extensive process disturbance. Moreover,
the state of the system may not be known exactly before making the decision. A
recommender system presents the user with the information that they are
interested in. The proposed recommendation system recommends the user with the
food that they prefer based on the user preference and type context. As there
are many food options that the user can consume, hence making them confused
about the foods available. Using the concept of Artificial Intelligence, we can
develop such systems for the ease of the people on the food recommendation and
also the use of AI concepts helps us for physically existence of such systems.
Recommendation system has got a great demand in recent years, and there are
several recommendation system in different areas as music, food, book, hotels,
news and more. Building an intelligent system which can suggest the food based
on user preference is difficult task. There exists several food recommendation
system with the fixed user preference, however the choice of user may change
depending on the context.