“Anytime you push some feature on your customer you really want it to be right for him/her 100% of the time.”
~ Steven Sinofsky
~ Steven Sinofsky
I had the pleasure to play around with the MEAN.io stack and it feels really easy to get started. I had mean.io up and running within 60 seconds, and i’m sure you can do it too.
For those who have no idea what the MEAN stack is, it is akin to the LAMP stack, but MEAN is built upon:
Here’s how you can get started with MEAN stack in less than 60 seconds:
npm install -g bower
npm install -g grunt
If you happen to see that there are no CSS related stuff loaded, than chances are the /lib folder are not found in the project. If this is the case, you need to run
sudo bower install --allow-root
first. Than, you need to run
again to make sure everything is loaded perfectly.
Now you have completed installation for mean.io.
And the 6 posts are:
I was browsing the web and I came across : https://flipboard.com/section/the-social-networker-bic5Z6
It has a great video background for it’s landing page, which makes it visually appealing and interesting.
Here’s a screenshot:
The main idea is to enhance user experience by reducing the number of clicks required for a user to reach her desired product page after a user performs a query.
The problems I am trying to solve are:
The solution i’ve implemented is to allow user to perform “drill down” during the query phase. Instead of just showing autocomplete suggestions, I show possible categories/attributes based on the keyword ( or product ) the user is searching for. The user can than select categories/attributes directly while performing search.
Feel free to view the demo (interface is in Mandarin Chinese) here:
This is the desktop version: note that as the query selects a brand name from the autocomplete suggested terms, the query suggestion switches to provide category/attribute selection.
Mobile version. This is a proof of concept done using powerpoint and some prototyping templates.
“Social Behavior of Investors” is my master’s thesis completed just a few months ago.
The problem I am trying to solve here is to matchmake investors with companies ( especially early stage startups ) in order to get the investment funds flowing. I did not focus on public companies as the data are generally public; a great amount of research has been done about them already. But what about the smaller companies where little financial information about them are available ?
The solution is to make model this problem as a classic link prediction problem: given a social network of investors and companies in Time A, predict if (which) new links ( investments ) will form between investors and companies in Time B. I’ve used dataset from Crunchbase for the purposes of this research.
The result is encouraging: the accuracy rate is up to 90% with AUC ( area under curve ) of up to 87% for both datasets that I have used.
A video demo:
 I derived 2 datasets from Crunchbase: using Facebook as a seednode, i collected companies, financial organizations and individuals 4 hops away from Facebook. The second dataset is based off RenRen, and collected companies, financial organizations and individuals 4 hops away from RenRen.
GoSpread.me was a weekend hack project when i was trying to help my friend market his Facebook fanpage and his website.
The problem: faced with limited budget and marketing options, it appears that spreading the word about his FanPage and website through his friends is the best way to get Likes and quality visits to his site since his friends all know him. However, there is no easy way to reach all his friends on Facebook; making wall posts/public posts don’t really help due to EdgeRank.
The solution: I’ve built a tool that allows my friend to send personalized messages to all of his Facebook friends in an automated manner. Think of it as “Autoresponder Meets Facebook Messaging”
Visit GoSpread.me here.
The main idea of Querybox is to allow non-technical users to quickly slice and dice through social data.
The problem I am trying to solve is avoiding the use of dashboards: i want users to immediately get to the dataset/point they are looking at. In a way, dashboards can be confusing and may or may not surface the exact data users want.
The solution we proposed is to use a smart assisted query that allow non-technical users know the type of queries they can conduct based on the dataset they have. Using the assisted query UI, users can conduct queries without the need to know programming.
The main idea is to help developers ( or even semi developers ) to quickly query Taiwan’s Open Data. However, since the Taiwanese government do not have a stable platform for hosting its dataset, we suggested hosting the data through Yahoo Query Language (YQL) platform with support from Yahoo Taiwan.
Taiwan’s government data is ported and made YQL ready. I’ve built a smart assisted query interface that allows non-technical people write YQL queries to extract government data. [ Video in Mandarin Chinese. ]
Watch the video here:
PS: this project was done for Yahoo Hackday (Taiwan) 2013.