Monday, September 14, 2015

Shouldn't there be a new method to calculate popularity of programs and channels in media?

Please share your views at quora:

The popularity of programs is based upon TRP, which is based on the number of views. But the TRP does not work, media make use of human psychology to bring content that is bullshit but can increase the curiosity of people, so that people watch it, even if they are psychologically and morally harmed by watching it.

The reason is best described in this article: Deepak Mehta's answer to Why does the Indian media promote pseudo-scientific disciplines like astrology, past life regression, other superstitions and bogus conspiracy theories?

If the popularity of programs would be based on feedback, then I think media will be changed drastically.

Popularity should not be only directly proportional to no. of likes (i.e. no. of views or TRPs) but also, should be inversely proportional to dislikes. Also, there should be a website where people can give feedback to the programs, such as in youtube.

Also, can we provide feedback through a set top box? Is it possible to make a set top box that takes account of feedback?

Friday, September 11, 2015

How health care sector can be improved drastically in India by bringing technology?

You can discuss the topic in quora here:

In the era of technology, life can be easier if we start using technology in our health care sector. The computer has gone much smarter to take decisions, thanks to data science. So why shouldn't we rely on machine learning to predict things? We can study a patient in a very personalized way and give them proper treatment.

So here are the steps:

First, we should build a national website, which contains accounts of all the citizens linked with their Adhaar Card.
Every doctor and nurse should be provided with a tablet. In order to make use of a computer to make decisions and predict things, we need digitalizing data. Without digitalized data we can't do anything. So pen and paper should be scrapped as much as possible.
I also found that many are hesitant to use technology because they think it's difficult to learn. They say so because they have never tried to learn the technology. A student in Class I knows how to handle mouse, how to play games in computer. So why can't a person who is more than 25 years old can't learn it in one month. Even a most illiterate can learn to play games in 3 to 4 days.

Digitalization is also required in remote areas even.  Actually, they need the most. But it has a different problem, lack of electricity and Internet.  But giving tablet solves the problem of electricity. Regarding the internet, people can work with even 2G to submit forms. Make an app on form submission, enter the data and hit submit. The textual data are much faster to upload than images. We should also make the app work on offline even (not necessarily to work only in real-time) , so that even if sometimes there is no connectivity, the work goes on and final submission can be made by end the of the day. If some are scared of losing data, they can make a print out of the data if needed.

In data science, a prediction is done based on what data we give, and how it is trained.

So what are the data we need?
As I am not a doctor, I can't say it in very much in detail. I am just giving an Idea about it.
First is feeding the prescription on the website. Each and every prescription should be stored in the database and should be linked with Adhaar card. Also, all the reports of different tests should be stored in the database. So for it, we need a good research, and specific form field should be made for every test. What form field we choose is very important. The benefit will be, when a patient again goes to the doctor, we will know the complete history of the patient. So, a patient will get a more personalized care.

If we have data we can make many statistical reports, such as what are the medicines doctors are prescribing. How can it be improved? What diseases people are suffering most, in a most precise manner, so that effective measures can be taken.

Data science can also help even predicting type and state of cancers (e.g. malignant or benign) by looking at the images if it is trained properly.

So, benefits are limitless. Only we have to identify, how it can be used and what are our problems.

Also does it require a huge expenditure. No. The attitude of our politicians is what it matters.

Do visit citizen response to health care in my gov: Click Here.