Why is MiraWow called as "Intellegent advertising" ?
|Clients:||Business Units or Small Business Enterprises|
|Section:||TWO in total of 5 sections|
In a late night at a university, a student is trying to complete his project in the course Neural Networks Applications. During break-time, he watches news and videos. However, he feels annoying because advertising appears suddenly and stops his movies. The reason is that the advertising is not concerned to his desires . In this moment, he wants to see some interesting subjects such as food introduction (because he is hurry) or travelling adventures (because of boring).
An idea is appearing in his mind: “Why does not have any apps which can understand you?”. An app can listen, watch the owner and show the most suitable desires at that moment. This is a primary reason and mission that MIRAWOW will be born. MIRAWOW is a website/ or an application which facilitates Small Business Enterprises to advertise to their customers intelligently. MIRAWOW is designed on R.A.M.B.O full stack web development procedure and have three artificial brains named Mako, Maeve and Maya .
Benefits of using MiraWow are as bellows:
- For users (YOU or customers): MiraWow provides a mass information of business store and selects the most suitable advertising to YOU. An advertising is selected which based on many variables such as gender, generation, location and trend at that moment. It will also be tailored when any business unit have their update information on their channel.
- For business units (SMALL BUSINESS ENTERPRISES or Client): MIRAWOW facilitates business units to advertise to their customers intelligently. It enhances proportion their products to their target customers (or correct customers: “who want to buy”) and reduce proportion of wrong customers (“who feels annoying)”. As a result, their revenues can increase absolutely.
MiraWow means "__L O O K __ W O W__" in Spanish
RAMBO FULL STACK WEB DEVELOPMENT
How R.A.M.B.O full stack are created?
|Data storage:||Azure/ Firebase/ AWS|
|Back-end||Python/ R studio/ Stata|
|Deployed:||IBM Watson/ Azure/ AWS machine learning|
|Section:||THREE in total of 5 sections|
R.A.M.B.O full stack web development is created by Dr. M. Ha . During creating a platform of MiraWow, he reconigsed that web designed which is based on several commons full stacks such as React-firebase stack, MERM stack, LAMP stack, LEMP stack, MEAN stack, Django stack…These full stacks are covered a web development roadmap from font-end (client side) to back-end (server storage). The drawbacks of these full stacks are that they DO NOT mentioned any data scient models (such as econometric or machine learning). Therefore, a R.A.M.B.O full stack was invented to include from data science models to web full stack development.
Font-end program languages are React/ Redux which links to data storages (such as Azure blob, Firebase, AWS) and communicate to users (YOU) and clients (business units) . Back-end program languages are used Python, R studio to analysis econometric models (such as ordinary lead square/ logit regression/ GMM regression) and machine learning models (such as random forest, support vector machine, multi-layer perceptron). All back-end program languages are operate under Anaconda environment. Back-end models are deployed in iCloud (a cloud computing service) server and connect to font-end through Nodejs .
As a result, three machine learning models are deployed to manage full MiraWow (from font-end to storages). These are Mako, Maever and Maya
Team members: MAKO - MAEVE - MAYA
The heart : Understand YOU
MAKO: The heart
This is a hybrid model which is combined between econometric and machine learning models. First, cluster analysis is used to visualize data and looking any clusters existing. Self-organisation map technique is used in this pre-step. Next, econometric models are multiple linear regression, logit regression, GMM regression. These models are used to investigate significant variables. Insignificant variables are considered as noise and removed from original dataset. Last, remaining variables are trained by machine learning models. These machine learning are random forest, support vector machine, multilayer perceptron. The final model is optimized and is deployed in iCloud.
According to Dr. M Ha’s study, his hybrid method is increased classification rate to 1.2%. In specific, Type II error (a customer feeling annoys) is decreased from 50% to 60%. This is a major reduction because this model will reduce number of unsatisfied customers significantly (50% - 60%).
MAEVE: The eye
This is a machine learning model using Python. The OpenCV/ COCO/ or Mobilenet libraries are used to analysis a feeling of users (YOU). If you are sad/ happy/ angry or …nothing, these emotions are considered as input factors for MAKO to find an optimized advertising which is the best wanted.
MAYA: The ear
This is a last member of the machine learning team. This model is also used Python programme language to design. The Tensorflow is used to analysis customer’s voice and convert into “key words”. These key words are used for MAKO to search a most fitting advertising which YOU care most.
As a result, three brains MAKO – MAEVE - MAYA are worked as an intelligent team to facilitate business unit advertising intelligently.
Coming Soon ...FREE DOWNLOAD
MiraWow app is avairable to download for FREE since JUNE 2021 at ....
Man's Invention Real ActionThe business system has seven business lines
Intelligent AdvertisingTo advertise intelligently
Cryptocurrency MiracoinCryptocurrency is spent arross contries
Mira international journal of avatar interligentThe journal of machine learning in Finance
Mira intelligent agricultureOrganic agriculture
Mira Intelligent HouseHouse for rent
MicruiseMira intellegent cruise
MislandThe longest journey
MIRA is a business system which includes seven business lines, includes:
- Online : MIRAWOW, MIRACOIN, M.I.J.A.I.
- Offline: MAGRI, MIHOUSE, MICRUISE, MISLAND