- Relevant topics
- Latent topics
- Informal participation
- Trends, clusters
- Actor mapping
- Sentiment analysis
- Real time analysis
- Ubiquitous analysis
Current challenges of informal participation
When we ask ourselves “what can Blockchain do for” or “what can Artificial Intelligence do for”, it is easy to begin with the solution (Blockchain, Artificial Intelligence) and then see whether that solution fits onto the “problem”.
In Current challenges of online participation: a citizen e-participation journey I listed what were the main hypothetical steps that one should follow to participate online, and what were the expected barriers or problems to overcome. The tacit idea being how could (if ever) decentralized ledger technologies or Blockchain help all along the whole way.
In order to prepare a meeting on Artificial Intelligence at the Administration and, most specifically, Artificial Intelligence for citizen participation, I here plan to repeat the aforementioned exercise, now for knowledge management in citizen participation, and most especially in informal participation settings.
Informal participation begins on formal settings. There are lots of things happening around the formal discussion of a given issue. Thus, it is not always easy to tell the relevant topics from the irrelevant ones. And in two ways: relevant for the organization of the deliberative process, and especially relevant for the participant, which may find burdensome to go through all the information and proposals and comments that everyone else is doing — this is especially true on online participation platforms.
Being able to distill what information is relevant (a) in relationship with the topic at stake and (b) in relationship with one’s own interests is crucial for the smooth and effective evolution of the deliberation process. Scanning, analysing, tidying up, summing up, and presenting in a clear way is the first demand to do to any technology aiming at crunching information for good.
In the same train of thought of the previous point, but now in more positive or constructive terms, there is a lot of hidden information, or tacit or informal participation, within a formal setting. E.g. a debate on urban mobility can tell a lot about the participants: their wealth or income, the place they live and their commuting possibilities, their jobs, their educational level, etc. It can even tell us a lot about other latent variables or issues such as environmental awareness, concern on pollution, etc.
Beyond the explicit message that participants are sending, it is very valuable to be able to extract much more information either about the profile of the participants (without damaging their privacy, of course) and other topics that may be on the public agenda but that the Administration might be overseeing.
So, formal participation can bring forward topics that were not on the agenda, or even provide evidence on topics or general information on issues that, without being made explicit during, do lay in the background of the deliberation.
But of course lots of things happen outside formal settings. People do politics in their daily lives, constantly. Most of them will never get close to an institution; some of them will even circumvent institutions as much as possible. So, getting to know where do these daily politics happen is something that the Administration could use to approach citizens where they gather. This is not to be confused with chasing people to bother them. On the contrary, it deals about knowing the ideal settings, the adequate code, the relevant question at the relevant time.
In a face-to-face world, and most especially on smaller towns, this information quickly spreads word-of-mouth and is well known by everyone. In bigger communities, and in the most liquid ones of the digital space, either takes years of personal involvement in several communities, or an artificial intelligence can come handy in making time shrink.
Artificial intelligence does not only shrink time, but can help to make the (many time hidden) connections for you through trend analysis: finding communities, finding their interests, knowing where clusters generate due to increased interest in one topic and, finally and from a dynamic point of view, knowing when a threshold is reached and there is a critical mask ready to take action or prone to follow one.
Beyond knowing whether there is a (hidden) hot topic on the public agenda and how it is evolving, it is very interesting to know what are the people interested in the topic and why.
It is important to note that what is interesting is not the specific identity of the individuals following a given issue, but their profile. First of all, because it may help to identify correlations between topics and, much more relevant, how intersectional a given topic is and the multiplier effects that take place when several issues overlap one another. E.g. we know that being migrant has a different impact on one’s life depending on whether the migrant is a man or a woman.
Identifying actors contributes to the necessary actor mapping that everyone should perform, thus, to measure the potential people impacted by a political instrument and, in consequence, to be able to formally invite them to any sort of formal participation initiative that the Administration may consider to foster.
As we have already said, the increasingly complexity of topics, the proliferation of shifting virtual spaces, the possibility of be part of different communities and play in different arenas, make it increasingly difficult that these tasks can be carried on by human beings, not to speak about a single person or a very small group with little presence on the streets and civil society organizations.
If we want to find people —actually their profiles and characteristics— it is but natural to associate their thoughts with how they feel about them. Are they speaking positively or negatively about an issue? Assertively or full of fear?
Real time analysis
Last, but not least, we want to know it now.
This may be not as useful for the Administration —that can take its time to sit and analyse— than for citizens at large. They usually have much less time to deal with public issues than public servants or policy-makers. Basically because they have to earn a living, which usually is not related to policy- or decision-making. Being able to grasp at a glance the state of the debate —most relevant information, main positions or approaches, types of actors dealing with the issue, most important arguments made— saves time and efforts and contributes to drive the deliberation to the most fertile grounds.
A a well trained artificial intelligence sure can help in providing this analysis, inference and main insights in (almost) real time.
Last, but not least, the application of artificial intelligence to support and improve citizen participation should not be limited to the possibility to feed it with documents. Sometimes data can be already produced in a structured way, sometimes it will be the most unstructured and natural of languages. Most importantly, most of the times the sources of information will be a mix of structured and unstructured information, different types of support —including image, audio and video— or meta-data coming from the interaction between humans and other machines.
In a nutshell, artificial intelligence can help citizen participation to make visible everything that was not, everything that was happening outside of formal spaces, everything that was happening even outside of our own conscience.
If you need to cite this article in a formal way (i.e. for bibliographical purposes) I dare suggest:
Peña-López, I. (2020) “Current challenges of informal participation: what can an Artificial Intelligence do for citizen participation” In ICTlogy,
#202, July 2020. Barcelona: ICTlogy.
Retrieved month dd, yyyy from https://ictlogy.net/review/?p=4762
Previous post: Three challenges of citizen labs