Some of the greatest promises that can come from this data collection include better insight into debt, credit scores etc. Algorithms that are used in social media companies can be applied to impoverished persons that help connect them to resources just as an advertisement would be matched to a user on Google or Facebook. Other studies have shown that collection of international phone calls in African countries can, throught satellite images, can show locations of high childhood malnutrition where additional humanitarian aid can then be sent. One major promise is the improvement of public health intervention, for example, in instances of a national health crises where mobile data can show areas that have the most devastation. Studies show that it could be possible for this data to reveal the effects of a natural disaster.
One of the major pitfalls regarding the realm of data development is regulation. There are few checks and balances that exist and private companies are able to collect information on users. Your location, name, age, the likelihood of the type of job you have, your income etc. can all be sold by websites and companies. Companies can say that they will not collect data on users without their permission, but it is likely that they will do it regardless. A set of regulations on data collection will vary across each country, thus making it hard to keep companies accountable across the board. If companies start leveraging information on citizens of one country that may cause them harm, it is very difficult for another country to regulate what goes on outside of their own territory.
Key steps that Blumenstock outlined as ways to look forward are aligning new data sources with pre-existing tools. He describes a “two-pronged approach” that assists data sources in complementing each other, rather than replacing them. Another way forward that is described is ‘customization’ in which specific algorithms are flexible and can be used in different ways. Another important note in this path forward is collaboration between all parties involved in data development such as “data scientists, development experts, governments, civil society and the private sector.”
“Good intent is not enough in data science when dealing with the problems which determine people’s experiences.” I like this statement because I think it encompasses one of the most important aspects of data development and that is the privacy and experiences of users. Mass data selection has the potential to greatly affect the experience of users and therefore should not be considered only with good intent, but with strict regulations.
Transparency is the underlying issue to many of these problems, so an increase in this on both ends (data based issues & human based issues) could lead to better results.” I think that this data collection should be a choice every user gets to make and I completely agree with this statement that increased transparency will lead to better results because people will be more likely to come forward and release that information.
“In lieu of such drastic potential for promoting applications yet demoralizing hinderances, the balancing act can become difficult.” This connects back to one of the biggest pitfalls which is regulation and the issue of balancing data collection and privacy.