Recall One Or More Situations In The Past Few Years Where Experimental Data Eventually Led To The

Thank you for this detail question description.The same things occur with every aspirants.They all do the same for fighting it off.But, what actually makes a difference is their unstoppable efforts to work onto the same.Fear, anxiety and pressure like things are very common in the struggle of clearing any competitive examination. Around millions are giving and thousands will be selected. Tough.Nothing except for going along and studying is the way to fight this.If you ask Google it has techniques like meditation and etc. If you start meditating with such kind of chaos in your mind then, you'll feel more low as you will depend upon meditation. Proper mediation gives result in and after 3–4 months of practice and till that time you don't have so much patience. Accept it. So it cannot help.As I said earlier, walking through the same. Studying is the way only.Hormones may play with you sometimes. Let it. But do remember when the game ends, you should be the winner.Don't feel low and stressful.Believe in your efforts, the results will walk to you as it is meant to be.Few things I personally practice to get rid of such things is:Yoga. Preferably, kapalbhati pranayama for 10 minutes daily. You can see the effects of the same on the internet.Weekly dopamine, serotonin and endomorphin doses. In my case I like to eat, so I prefer eating tasty and healthy food every week to keep myself happy.Build up the habit of patience. Slowly everything fades off and I'm again back on track.Discuss the issues with one close friend and with family. I said discuss, don't expect solutions. It will help.Finally, be passionate and keep reminding yourself, your 'why' for this journey.All the best!Thank you!Binoy Roy

There are various ways to measure the relevance of a search engine. In general, instead of the standard machine learning "relevance" metrics such as precision, recall, accuracy, AUC etc., search engines prefer a different set of metrics that I will highlight below. The non-A/B test way to measure relevance is to use human raters and ask them to rate the results for certain queries. These ratings are then aggregated across a query sample to get your final relevance metric. NDCG is a common metric used in human relevance systems like these. I go into greater detail about this technique here: Nikhil Dandekar's answer to How does Google measure the quality of their search results? But I assume in this question you are more interested in relevance metrics that can be measured in an "online" A/B test. In an A/B test, you show a different ranking to different cohorts of users and see which cohort shows better overall engagement with the search engine. In an A/B test, you have to rely on user actions to measure relevance. On a per query level, the most common and easily measurable user action is a click. Apart from clicks, there is also derived data that can be extracted from the clicks such as time to click, time between clicks, position of the click etc. The most common metrics used to measure relevance using this click-derived data are:Clickthrough rate (Simplest, but has a bunch of issues)Mean Reciprocal RankPrecision @kClick-based NDCGetc.At an experiment level, you can use other, indirect metrics to measure engagement such as:Number of user sessions: You want users to your search engine more, so you want this to go up.Queries per session: You want the users to get to what they are looking for faster, so ideally you want this to go downAverage time to success: Same as above, you want this to go down.In general, your success metric should be tailored for the what your exact A/B experiment is.Finally, you can also do result interleaving, which is a way of doing an A/B test at a per query level, so that it converges faster. I have written more about interleaving here: Nikhil Dandekar's answer to How do I evaluate quality of search results for different algorithms or enhancement without AB testing?

There have been many situations in my life which can be classified as difficult.One in particular was a period of severe impoverishment (I lived in another country at that time, during my childhood).Yet, there is a caveat: some situations were not under my control (that same period of impoverishment and deprivation) and others were under my control (and at times, I even caused them).That latter part was the biggest obstacle to overcome.Not that I caused those situations on purpose but it was a pattern. It took quite a few years to recognize that it was indeed a pattern, and that it was my choice that was the reason why same things kept on happening over and over.In a moment of realization that I would be able to change the situation by changing my attitude towards it, was a breakthrough.I actually became a more aware person basically overnight (with all those years in the making).It actually started with a catalyst - a child.I had a child, and I don't know whether it was a true enlightenment or the sleep deprivation, but at some point I had to put imaginary words to those sounds that the infant was making.I used to be a cheerful person and a witty one - but with the speechless infant, my humor superpowers were utterly useless: crying means distress, and it's quite depressing to hear it over and over (early in life, infants don't giggle, so it's either crying or silence. Take your pick).So, after some internal struggles, I had to change my ways and accept that I needed to be more patient.From patience, came acceptance and I became more observant. My immediate reward was not to panic when the child was crying (yes, it's a huge gain for a newbie parent) and first, seek to understand.Once the internal need to be understood was superseded by the need to first understand another, miracles started happening :)