A Visualization For Score-Cutoff-Like Strategies

It’s been three straight months since I started my work at HomeCredit China, and I have to admit that companies like this big really take new employees a lot of time to understand their business model at the beginning. But thank god, time and intelligence can together help me to overcome the difficulties, so now I am starting to get the hang of it, and I am going to share some of the techniques I found out in my work that can help others, potentially.

Working in the decision-making department that controls the underwriting strategies for loan applications, quite often we would use “scores”. They could be scores calculated by the company it self, or the score from other companies like the popular “Zhima Score” from Alipay in China, or they could be a little bit of either. For example, if we were Alipay, and we are gonna say in order for us to credit you some money, you’d better have a Zhima Score that is higher than 590. One day you found out that this decision is rejecting more customers than you would hope for, you’ll naturally want to adjust the cutoff value, right? Therefore you’ll start to simulate the new cutoff value, for example, 580, to see if this new strategy is doing you any better. Simulation is also important but I’m not gonna talk about it here.

However, what I just said is just for the most simple case, where you only have one score cutoff, so that you can just rerun the simulation again and again to draw a graph where the x-axis shows you the cutoff value and the y-axis shows you the approval rate, for instance. What if, you have a lot of score cutoff strategies? You can still do the analysis one by one by rerunning the simulation by different cutoff combinations, but when the data for simulation is large this process can become crazy slow, and you’ll not be able to see their interactions at first glance.

What I am proposing here is, to use an “incremental” method for visualization instead of the “all-over-again” method. Let’s get the simulated data once, and then do the modifications on it. For instance, when you want to lower the Zhima Score cutoff, try to find out what part of customers in the dataset will be possibly getting the offer, and then assign a possibility of them being approved, such as 80% or any other value calculated. Correspondingly you can apply this modifications to all other cutoffs, and because the strategies in the decision-making system are usually run in an order, and are coded in a table with each line referring to one strategy, you could break down the specific segment of customers who might get new results, calculated by probability. It is not accurate at the first glance, but when your data is large, it could be more precise than you would’ve thought.

And the huge upside of this method is, you could visualize the interactions among several score cutoff categories AT THE SAME TIME! You can play with the modifications and see their general impact on the variable you’re interested in. In my example here you can play with the sliders for different score cutoffs, and see what they would do, how large the impact would be for the approval rate and risk performance. At the very end, when you find out the perfect combinations you want, you could simply rerun the simulation to prove this thought.

And trust me, the result won’t be too far away from the visualization, because you already know what is going to happen through the visualization.

Project link here: Github Link

Signing up the Dataquest!

Since I started to work, the things to do have mainly become getting data from database using SQL, and then throw it into Excel to do some analysis – to be honest, Excel is enough for most analysis for most companies. Although in fact my supervisor has never constrained his employees what kind of tools to use, most of my colleagues have settled down on Oracle SQL + Excel.

This is not the case for me, never. As one of the only two people who can code in my team, I started to automate some processes (especially some annoying and repetitive ones) using Python (thank god I have learnt Python myself) with another colleague who is from computer science background. It is pretty funny when he also learns the fact that I could even code in HTML and CSS, and I believe he would never underestimate a business student’s coding skills anymore. Another interesting thing about my job is that one lousy colleague always keeps asking questions about R programming language to me. One day when I tried to asked her some very basic operational questions, she was pissed off for no reasons and shouted “why don’t you check it up yourself”. After that I’ve never talked with this idiot who thinks she is the center of universe and have rejected all possible chances for conversation. Please, Chinese girls, be respectful to others, will you?

Alright, back to the topic about continued learning after graduation. Apart from my learning Japanese everyday, I sincerely believe that I need to keep learning some professional skills to make myself competitive for the next couple of years. Tensorflow is what some others are learning, and I took a look at it. It is a very promising package from Google but, it is too cumbersome for learning and coding. Not long after Keras bumped into my vision, and I think it will be a perfect package for deep learning beginners. After doing some research online, I happened to find out two amazing websites for people who want to learn more about data science: Datacamp and Dataquest.

I really want to try both at the same time, but they both require subscription to be able to access to all available courses. Datacamp is more R-focused with growing Python content, while Dataquest is more Python-focused with clear paths to be an expert in data analysis in Python. No brainer, isn’t it? As a guy who has learned R extensively in the past two years, I’ll be more than happy to have a website who specializes in teaching people Python to do data analysis, so I ended up starting my learning in Dataquest. I might be signing up to Datacamp later as well, but it depends on how fast I can go through the content on Dataquest (I am learning like flying because I want to save some money).

No matter which service I chose, my ultimate goal is to get my hands on deep learning after gaining some machine learning knowledge from USC. I believe being able to know it as well as to do it, will greatly enlarge the boundary of who I can become. I just want to be better by learning things that interest me, and I know I will.

Thoughts on Ulysses changing to subscription-based

I have been using Ulysses for quite some time and consider myself one of the first group of adopters for this application. Ulysses feels elegant, fast and effective, and has been my preferred writing app.

When Ulysses announced its change from a one-time-paid program to a subscription based app, I was not shocked at all, since in recent years 1Password, Adobe Photoshop, PyCharm and other excellent softwares have all changed from free/paid to the subscription type. It is not necessarily bad for consumers at all, if this move can help developers to continue its work for a long long time. On the other side, it does have added extra financial pressure to their clients if customers need any new feature from the app, because almost all developers will give no time to customers to think about it and release a new version directly, at the same time taking down the old guy from the store directly.

Honestly speaking, I don’t think customers who have paid such a high price for such an excellent app would be mad at the decision to switch. As one of them who might be unhappy about the move, I am more unsatisfied with their incompetence in notifying existing customers beforehand. I think, even after nearly 10 years since the concept “App Store” has been created and implemented, we are still searching for the best way for developers and customers to benefit enough from its booming development. Probably those platforms are taking out too much from the developers? Or customers are paying too less for some applications? Only time will tell.

Life is a Journey of Accumulation

Sometimes I wonder what makes me today, and there’ll be tons of alternative answers in my head. It might be due to my interest, my characteristics, the environment that I grew up in, and the people I have met in my life, or it is simply due to the destiny.

For example, I’ve been studying wine culture and Japanese recently, but I found myself constantly forgetting the new things I just memorized. I’ve never wanted to blame my seemingly decreasing learning skills, but I found that to be good at something, it takes some serious time in practice. To be accurate, I have spent quite some time seriously in learning the things I am good at today, instead of doing the things repeatedly without ever wondering how to improve it.

It comes down to a book talking about “deliberate practice” that I read in the past couple of weeks. Looking back to my journey as a tech lover, I did also encounter some time when I found myself keep forgetting new things that I just studied. Gradually, however, I started to pick them up more and more quickly, and that’s the time when I realized I had accomplished something in learning a new thing. So my point is, don’t see yourself as a slow learner, because we have never been that fast. The reason why now we are good at something, is that we have spent countless time in mastering it, whole-heartedly. We tend to forget how much efforts we have put into something, especially after we are sort of comfortable in using it.

There’re just too much great stuff to learn in the world, and I used to be too keen on doing all of them. After all these years, I’ve also learnt to enjoy myself a little bit when it is time to do so. And no matter what I want to do, good or bad, hard or easy, I’d like myself to remember the art of patience in learning, since it is better late than never. 🙂

DotA 2 – A Modern MOBA Giant

I am never a hard-core gamer before. Athough I do have ever played good RTS games like StarCraft, Red Alert 2 or Empire Age, as well as others such as the famous FPS game – Counter Strike, I have never spent any second on Multiple online battle area (MOBA) games before. I can still remember clearly the time when all of my friends were crazy about Warcraft and Defense of the Ancients, because that was the first time I knew that some games are really popular, but I didn’t try it, until recently I got my hands on the DotA 2.

First things first, DotA 2, as a MOBA illustrator, is impossible to play on mobile devices or game consoles that without a mouse, so it has only been released on PC and Mac platforms. Similarly, the well known Riot Game’s League of Legends is the same, being accessible only on computers. The reason for that is because these MOBA games are so in depth, with tons of options and settings, and requiring a lot of accurate clicks and drags throughout the game session. Typically, a DotA 2 session can go from 40 to 50-minute uninterrupted time, with a 5 V 5 competitive gameplay. Thus saying, DotA 2 is also a team play game. You’ll need teammates to cooperate together to destroy the base of the other side, and vice versa.

It is an extremely difficult game to get hands on, since its depth in gameplay. But once you are in, you are in for life. If you are a casual gamer, it might not be the one for you, but if you are a purist, you have very little room to complain about its excellence in practice of MOBA. Being such a deep and well-balanced game, it is not without drawbacks. For example, it is too “harsh” on newbies, and it is also not mobile-welcoming. The complexity it possesses is a gem but also a fallback. It also lacks in the customization in skins and items, just like what Apple does to its iOS, limiting what the gamers to do to change the looks in the game, while instead offers exclusive skins on sale to attract hard core fans.

That said, it is a great game to be learned from if you are interested in the market of MOBA games, but it is also important to keep its dissect simple to not let readers down. Analyzing DotA 2 truly could be written into an entire book with tons of details and complexity.
Continue reading

Main Criterion to Buy a Gadget/Service/Apps/…

There’re tons of things coming up everyday in this fast-growing technology world, and customers like me have tons of choices when it comes to buying any kind of stuff. Then what is my standard of it to help me avoid lots of bad things out there on the market?

Needless to say, when you are facing some choices or alternatives in a specific type of service, for example, a GTD app on your smartphone, typically on the iOS platform you’ll see OmniFocus 2, 2Do, Wunderlist and so on. The choice of mine is OmniGroup’s OmniFocus 2, an easily made decision for me. The reason why I chose this app/service is because the company and history behind this app. OmniGroup has been a excellent and consistent developer for both MacOS and iOS platform for over 10 years, and I have no reason to choose other apps when I know I can get a steady service from a more trustworthy company in my eyes. When you buy an app, you are not only buying the service now but also the service in the future, since nobody is hoping to get an app that will be taken down from the App Store or discontinued by the developer. For instance, when the app Ember from Realmac was discontinued, a lot of customers were whining about the developer without giving enough efforts to loyal customers, with a fact that this was an expensive and professional app. If people spend $60 in an app but only get 1 year of use, then it is totally unacceptable, making the developer’s other apps lose trust from customers.

To take this discussion further, for instance, what if you want to buy a car? The first thing you need to consider is the company and history behind the brand, again. Take Volkswagen as an example, the diesel car scandal truly has made some customers distrust this company anymore, but the history behind this brand has still enabled itself to attract new customers worldwide. Another example being Samsung, we all know what happened to its 2016 flagship smartphone Note 7, but somehow its strong background and good history have together helped make a comeback this year with the new S8. It seems like I am persuading people to buy "old-fashioned" stuff instead of trying new things, but actually I am not. I am, too, a guy who likes to try new things, for example, I went to grab a Nintendo Switch before I can’t:

If we also want to touch on the criterion on buying/trying new things, it is also easy to tell: Also buy those things that have overwhelming positive comments and reviews. Sadly Nintendo Switch is one of those great things, so before I hop on the flight back to China where I can’t get one as easily as I do here, I decided to grab one from Amazon last week when it was available. Okay, well, perhaps I slapped on my face regarding to what I have said in my last article, but NS is so damn good! Another good exercise for my standard lol!

Especially in the tech world, things like kickstarter or beta projects from some companies are selling concepts or ideas only to customers. Things being purchased might not be worthy of owning. If you are a tired customers like me, buy things that have been well tested and are from a well-known company. If you want to try something new, buy things that have almost one-sided positive image. Any final words…? Yes, please avoid flying with United Airways, it is just not a reliable company.

Buying a Nintendo Switch?

The first two weeks since the launch of Nintendo Switch has been unprecedented, with Switch already being the most popular game consoles that have ever been created by Nintendo. I, as well, am quite tempted to get a Switch despite of its rarity at this point, but there’s also a voice inside of my heart telling me to be calmed down.

The blockbuster, Zelda: The Breath of The Wild, is a 10/10 game according to various game-reviewing websites. It is quite amazing in details and controls, and for sure is an awesome game to play, locking in the best game of the year award. However, this game has also pushed the console powerhouse to the limit, depicted by quite a lot of people, so I can easily assume that more power-consuming games, such as the upcoming FIFA and NBA 2K18, will not be as amazing as their PC/PS/XBOX versions. Nintendo is not famous for the video quality of its games, but for its fun instead, so it makes sense by limiting Switch’s resolution to 720p regardless of the current trend in pursuing pure horsepower.

After all, apart from Zelda, there’s no other single game that is quite convincing for gamers to buy. It’s not because of the lack of performance of Nintendo Switch but the lack of third party games that makes it not quite worth buying right now, unless you are a Nintendo fanboy. I do appreciate what Nintendo is offering on this platform, and I also believe Switch will be one of the best consoles in the future, but I will choose to wait for several months to see how the other games and companies are planning to follow up on this platform. Wii U has been a disaster in third party titles, and making itself uncompetitive to a lot of gamers. The mainstream type of gamers, as it is displayed, is the one who play PC/PS/XBOX titles around the world. If Nintendo is going to make Switch a worldwide phenomena, it still has to prove it.

Seriously, I am going to get a Switch someday, but not now, not near.

The Twitch Generation

I hate to say, that about two months ago, I was one of the group who claimed that people who love live streaming shows are idiots. I sincerely apologized for my mistake, because if that is true, I am the biggest idiot since now I am one of them.

Strangely enough, the live streaming platform is, sort of, lead by asian platforms. I know Twitch is probably the most well known in America, but after browsing through nearly all contents available, I have to say, asian countries are the kings nowadays, especially in South korea and China. On Twitch, live game streaming is the mainstream, and so is true for Korea and China. However, asian streaming platforms do so much better in other miscellaneous contents such as karaoke, pretty girls, eating, outdoors, and so on. YOU NAME IT.

Plus, those platforms even have ambitions to enter the entertainment market, by which I mean, they try to cooperate with TV stations, so so co-op shows can be broadcasted on TVs, making more and more people realizing "Oh, this platform has nice shows!"

The lengeday hot pot of Korea, eating LIVE!

Continue reading

After all this time, has iPads been “Pro”?

Approximately 1.5 years ago, Tim Cook announced the "best ever" iPad that Apple has ever made — the 12.9 inch iPad Pro. Up until now, this gigantic piece of glass is still the most relentless powerhouse among all IOS devices and even in the all tables market. Recently I have listened to quite a lot of podcasts of Canvas and Mac Power Users in the Relay FM programs, and I sort of find out the unsatisfying fact to all Apple fanboys that iPad is far from a PC replacement, and it is due to the software instead of the hardware of it.

Around two years ago, the IOS 9 has marked a huge improvement in IOS productivity field, by adding the share button and some iPad-only features like split screen. Continue reading

The Grand Tour – Legends never die

About three weeks ago, I bumped into a video on Youtube, and it was about one of the three original hosts and probably the most important one, Kai Man Wong, from DigitalRev TV which is a Youtube channel that talks about trends and trips for cameras and based in Hong Kong, has officially announced his leave from the channel. It was a destroying news for the fans, following Alamby’s leave (another original host) about three to four years ago.

Kai, who is a great host in the area of cameras, once lead in the production and script writing of the channel, was mysteriously headed back to Britain about one year ago, and since then I have sensed the potentials of his leave, sooner or later. What surprised me, though, is not his leave, because I have already skipped a lot of the videos of that channel since without him the shows from the channels just don’t look complete. What surprised me is, soon after his announcement, the cameraman of the channel, Lok, also announced his decision to leave.
Continue reading