Try looking at ways where you can use machine learning on the data generated. This will help analysts to have reports they never thought of ie by using mining algorithms that can learn to identify training samples not intially labeled and thereafter model sensible information or analytics out of them without user interventions/ explicit programming. BR, Wilson. On Tue, Nov 1, 2011 at 4:43 PM, Leonard Kore <leonardkore@gmail.com> wrote:

Hi guys, I am a student collecting ideas for my fourth year project.Which apps do you think I can create using the information on open data.

_______________________________________________ Skunkworks mailing list Skunkworks@lists.my.co.ke http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks ------------ Skunkworks Rules http://my.co.ke/phpbb/viewtopic.php?f=24&t=94 ------------ Other services @ http://my.co.ke

-- Kind Regards, Wilson B.

Allow me to NOT answer your question. The way I see it, you probably have 3 objectives: 1. Score an A in your project 2. Win some award in some competition 3. Get offered a job soon after graduating Working backwards, the goal will determine the "type" of project you might want to consider doing. 3. Get offered a job Do a project that solves a REAL problem in the REAL world. Do not sit at your computer and imagine what problems exist out there, instead, read the newspapers, watch news on tv and find a real problem that afflicts someone or some organization, then do a project around that. 2. Win some award Go ahead and go a gisty sounding project using some buzzword technology stack, you can use the open data if you so wish. Google organizes lots of competions; summer of code, code jam etc and so does microsoft, e.g. imagine cup. Samsung and et al also organize competitions based on their mobile platforms. 1. Get an A in your project Come to think of it, point number 2 above will most likely win you an better grade than point number 3. But remember, most likely if you go looking for a job from Rad! et al, most of them will care less if you got a A in your 4 year project. Finally, last time I checked, it was not possible to buy stuff off Nakumatt's website and pay with mpesa and get them delivered to your preferred location, for a multi billion company, that a glaring oppportunity :-) On Tue, Nov 1, 2011 at 4:43 PM, Leonard Kore <leonardkore@gmail.com> wrote:

Hi guys, I am a student collecting ideas for my fourth year project.Which apps do you think I can create using the information on open data.

_______________________________________________ Skunkworks mailing list Skunkworks@lists.my.co.ke http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks ------------ Skunkworks Rules http://my.co.ke/phpbb/viewtopic.php?f=24&t=94 ------------ Other services @ http://my.co.ke

And read this: http://www.kalzumeus.com/2011/10/28/dont-call-yourself-a-programmer/ On Tue, Nov 1, 2011 at 4:43 PM, Leonard Kore <leonardkore@gmail.com> wrote:

Hi guys, I am a student collecting ideas for my fourth year project.Which apps do you think I can create using the information on open data.

_______________________________________________ Skunkworks mailing list Skunkworks@lists.my.co.ke http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks ------------ Skunkworks Rules http://my.co.ke/phpbb/viewtopic.php?f=24&t=94 ------------ Other services @ http://my.co.ke

Here are some ideas. May be good or bad, just stuff I've thought would be interesting. 1. Geomapping Nairobi traffic: The static images provided by the Access Kenya webcams are difficult and cumbersome to use. Ideally, you should be able to use basic image processing techniques (object detection) to work out how busy specific junctions are and relay this information onto a livemap. This would *greatly* improve the functionality and impact of the system. I have specific mechanisms and code that will help you to extract vehicular information from the images if you choose to take this project. I even have routing enhancements to A-* to help with this. There are other ways you could improve this; for example, if you wrote an Android App that made use of phone accelerometers in conjunction with GPS, you could (roughly) estimate traffic flow and speed thus creating a crowd-sourced traffic information system. 2. Dealfinder Nairobi: I'm always on the lookout for a good deal, whether it's half price drinks or a 50% discount on TVs. I find in Nairobi the issue is not so much the deals (there are some good ones) but being aware the deal exists. A software infrastructure, possible relaying information over SMS that allows people (individuals) to notify each other of relevant deals would be useful (think of it as SMS based, niche-market twitter with groups). This can of course be generalised to "any" data, not just deals. (e.g. you could use it to work out where rolling blackouts are happening on a weekly basis). 3. While the data released on opendata.go.ke is useful, it's practically useless to the ordinary non-technical citizen. What would be really nice would be a website that visualises this information in a semi-coherent manner, allowing citizens to make interesting queries and/or determine how the government is spending and acquiring income. This is a really simple idea, but if done well could be useful in empowering the ordinary citizen. 4. C me C u - It would be really cool to be able to use the (compute) resources of the people around me for my benefit, specifically for recording incidents. Often an event will occur and I'll want to record it so I'll take my phone out and start to take pictures. There will be people around me taking pictures as well, some are better than mine, some a re worse. I often end up sharing pictures with them on a one-to-one basis. It was nice if this was seamless - and with Android it can happen quite easily. 5. Clever News: I go to the daily nation, all african news, standard and various other websites to get my information on a day-to-day basis. This is cumbersome and can be annoying when I end up with repeat articles. A piece of software that can pull in news from various sources, analyse it and discard semantic duplicates would be pretty cool IMO. Note, this is totally different from RSS. Again, if you want to go down this direction contact me and I'll put you in charge with a person who runs such a site and could actually use your output. --rip On Tue, Nov 1, 2011 at 1:43 PM, Leonard Kore <leonardkore@gmail.com> wrote:

Hi guys, I am a student collecting ideas for my fourth year project.Which apps do you think I can create using the information on open data.

_______________________________________________ Skunkworks mailing list Skunkworks@lists.my.co.ke http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks ------------ Skunkworks Rules http://my.co.ke/phpbb/viewtopic.php?f=24&t=94 ------------ Other services @ http://my.co.ke