Hierarchical Attention Network for Visually-Aware Food Recommendation Java


Food recommender systems play an important role in assisting users to identify the desired food to eat. Food Ordering System is an application which will help the restaurant to optimized and control over their restaurants. For the waiters, it is making life easier because they don’t have to go the kitchen and give the orders to the chef easily. For the management point of view, the manager will able to control the restaurant by having all the reports to hand and able to see the records of each employee and orders. This application helps the restaurants to do all functionalities more accurately and faster way. Food Ordering System reduces manual works and improves the efficiency of the restaurant.

Deciding what food to eat is a complex and multi-faceted process, which is influenced by many factors such as the ingredients, appearance of the recipe, the user’s personal preference on food, and various contexts like what had been eaten in the past meals. Because of the importance of food in everyday life, food recommender systems have become an indispensable component in many lifestyle services, and are often touted as potential means to affect people towards a healthy lifestyle we formulate the food recommendation problem as predicting user preference on recipes based on three key factors that determine a user’s choice on food, namely, the ingredients of a recipe and the descriptive image of a recipe. To address this challenging problem, we develop a dedicated neural network-based solution Hierarchical Attention-based Food Recommendation (HAFR) which is capable of capturing the collaborative filtering effect like what similar users tend to eat inferring a user’s preference at the ingredient level and learning user preference from the recipe’s visual images. To evaluate our proposed method, we construct a large-scale dataset consisting of millions of rates.  Extensive experiments show that our method outperforms several competing recommender solutions like Factorization Machine and Visual Bayesian Personalized Ranking with an average improvement of 12%, offering promising results in predicting user preference on food.

Existing System:

Many Restaurants stores and maintain their day to day transactions manually. But some of them are having an automation system which is helping them to store the data. But such restaurants are storing information about the orders and customer information. They don’t have the facility to store the information of feedbacks and favourite orders of customers over some period of time. Restaurants are having standalone applications so at one time, they have the facility of many screens or many operations which is happening at one time. So they are storing them and then at last, the restaurant managers will able to see the data of last day.
The Internet and mobile technologies facilitate people to access information at any time and place. People’s lifestyles have been changed profoundly with the spread of online services, such as social media, E-commerce and various lifestyle Apps.
Because of the importance of food in everyday life, food recommender systems have become an indispensable component in many lifestyle services, and are often touted as potential means to affect people towards a healthy lifestyle.
It has approached food recommendation with a comprehensive consideration of all the above-mentioned factors. Existing work has either adopted collaborative filtering which is limited to user-food interaction modelling, or performed content-based filtering based on the ingredients or image features.


At last the restaurants have to store by themselves which will became no use of software.

The user interface of the application is also not that much attractive. So from the restaurants point of view, they are able to store only one kind of information. There is no security feature also.

If any of party order is cancelling at the last moment, it will make a variation in the already created records and also will causes the wastage of foods.

Many of the systems will not store the budget details for a long time.

Proposal System:
This system is a bunch of benefits from various points of views. As this online application enables the end users to register to the system online, select the food items of their choice from the menu list, and order food online.
Also, the payment can be made through online mode or at the time of home
delivery depending upon the customer’s choice and convenience. The selection made by the customers will be available to the hotel reception or to the person handling work assignment. It defines the payment to be done by the customer for order placed from the web store at worth price.
We proposed SVM algorithm used to classify the dataset into location, price, cuisine, veg or non-veg, delivery and feedback. To analyse the sentiment, we have proposed the collaborative filtering algorithm. This algorithm is used to detect the positive and negative feedback. The user given feedback is checked with positive and negative dataset by collaborative filtering algorithm.

These interjections are all too familiar for anyone who takes orders over the phone. Occasionally, a misunderstanding occurs or an employee takes down the wrong order. Cue the angry customers, wasted food, and disappointed manager. With online orders, the customer makes everything clear on their end. Everything is in online, and there’s no mix up. 
Offering online ordering lets your guests place an order more conveniently. Without feeling pressure to wrap up their order, customers are more inclined to explore all of their menu options, and even end up spending more than they would when ordering over the phone or in person.

Modules Description:
Upload Dataset
View Restaurants
Admin can directly login to the application. After login, admin has to add all the restaurant information’s into this application. we have used restaurants dataset, now to upload this dataset. Now Added restaurant location of the restaurant will be shown in graph. If admin wants to remove the service means, he can remove from the application.

2. User
Search Restaurants
First, user needs to register with all details, Then Login with username and Password. In this project, we have two type of search
1. By giving only the location
2. Price, Cuisine type, location 

First Search: In 1st search, user needs give the location and they can search for particular location.
In this search, Feedback of restaurants also taken in account. Based Upon the feedback it recommends the restaurants.
These are the restaurants available in Velachery. Top most restaurant is Chilli’s American grill bar. Because it has good feedback on comparing to other restaurants. Here user can choose any restaurant and order food with price. My Booked Items are added to cart. 
CarT: Here have make payment by using my card information’s After payment, the confirmation is generated to user s mail id.
Rating:  here user can give feedback for restaurant

Here we have to give 3 attributes to search the best restaurant
2. Cuisine type
3. Location
For second search we have used SVM classifications
In second search--> SVM algorithm will give high priority to the restaurant which has veg and non veg type, 
delivery, and then feedback, price, cuisine type.


System : Pentium IV 2.4 GHz.
Hard Disk        : 40 GB.
Floppy Drive : 1.44 Mb.
Monitor : 15 VGA Colour.
Mouse : Logitech.
Ram : 512 Mb.


Operating system : Windows XP.
Coding Language :  JAVA
Data Base :  MYSQL


[1] Home KFCKU. Available: http://www.kfcku.com. Accessed on June
15, 2015.
[2] McDelivery Indonesia. Available: https://www.mcdelivery.co.id.
Accessed on June 15, 2015.
[3] PHD Ahlinya Delivery. Available: https://www.phd.co.id. Accessed on
June 15 2015.
[4]D.S. Johnson, L.A. McGeoch. ”The travelling salesman problem: A case
Study in local optimization” Local Search in Combinatorial
Optimization, 1997, pp 215-310.
[5]F.A.T. Montane, R.D. Galvao.” A tabu search allgorithm for the vehicle
Routing problem with simultaneous pick-up and delivery service”.
Computers & Operations Research, vol. 33, no. 3, pp. 595–619, 2006.

Leave a Reply
Name E-Mail Phone