18 Bridging the Information Gap: Mapping Data Sets on Information …
223
18.5 Future Research
Our research focused on the relation between available data and information needs.
Although we inventoried Decisions, Activities, and Information needs, we did not
investigate the relationship between these three elements into depth and the system
dynamics between the different stakeholders including the political and financial
dimension. These dimensions played out, for example, in the still largely separate
data collection processes between NGOs and government and in when a flood is
declared an official disaster. Further research could address humanitarian decisionmaking in terms of “what may be influencing decisions, other than the needs on
the ground” (Nissen 2015). A political analysis of the stakeholders and the financial
flows might strengthen the information management research approach. Regarding
the relation between available data and information needs, it will be worthwhile to
determine the time dependency of the information needs into more detail and to do the
mapping on the data products in a more automated fashion. For large organizations,
it is possible to map through which information channels (email, mobile, fax, and
chat) information consumers get information products from internal information
producers. This kind of mapping does, however, not consider the degree to which
information needs are covered. Furthermore, it is much more difficult to do this kind
of mapping between organizations and even more so if certain workflows are still
paper-based. It might be possible to log data file usage on the main websites that are
used by responders and, for example, how the app and dashboard are used (Pachidi
et al. 2014). In addition, an after-action review with the responders in a focus group
setting could be used to have the responders categorize their needs according to
the four phases. This refinement could lead to an enhanced understanding of the
data gaps. We envision two avenues to further close these gaps. The first avenue
consists of assessing how Artificial Intelligence for Disaster Response (AIDR), such
as data and text mining, can be used to link and integrate disparate data sets and to
in this way reach a higher coverage of information needs (Spruit and Vlug 2015).
It will not be necessary to integrate all disparate data sets; we showed that a very
good coverage of information needs can already be reached by integrating the three
most important data sets out of the total 14. One could set up so-called data spaces
which are loosely integrated sets of data sources where integration happens only
when needed (Hristidis et al. 2010). This could become an essential extension to
the earlier mentioned data exchange platforms so that these platforms offer—to a
certain degree—sensemaking of all the data sets that are shared through them. The
second avenue consists of tackling the lack of local and timely data. The Government
of Bangladesh has started to develop an online process of collecting the SOS and
D form data, tackling in this way the data granularity loss. We have co-created a
smartphone application in Bengali that local disaster management professionals and
volunteers can use to collect data just before and during the floods that fulfills the
currently not covered information needs. The functions and features of the app and
dashboard reflect the different clusters of information needs that we identified. The
data collected is fed back to the affected communities through a dashboard that is
223
18.5 Future Research
Our research focused on the relation between available data and information needs.
Although we inventoried Decisions, Activities, and Information needs, we did not
investigate the relationship between these three elements into depth and the system
dynamics between the different stakeholders including the political and financial
dimension. These dimensions played out, for example, in the still largely separate
data collection processes between NGOs and government and in when a flood is
declared an official disaster. Further research could address humanitarian decisionmaking in terms of “what may be influencing decisions, other than the needs on
the ground” (Nissen 2015). A political analysis of the stakeholders and the financial
flows might strengthen the information management research approach. Regarding
the relation between available data and information needs, it will be worthwhile to
determine the time dependency of the information needs into more detail and to do the
mapping on the data products in a more automated fashion. For large organizations,
it is possible to map through which information channels (email, mobile, fax, and
chat) information consumers get information products from internal information
producers. This kind of mapping does, however, not consider the degree to which
information needs are covered. Furthermore, it is much more difficult to do this kind
of mapping between organizations and even more so if certain workflows are still
paper-based. It might be possible to log data file usage on the main websites that are
used by responders and, for example, how the app and dashboard are used (Pachidi
et al. 2014). In addition, an after-action review with the responders in a focus group
setting could be used to have the responders categorize their needs according to
the four phases. This refinement could lead to an enhanced understanding of the
data gaps. We envision two avenues to further close these gaps. The first avenue
consists of assessing how Artificial Intelligence for Disaster Response (AIDR), such
as data and text mining, can be used to link and integrate disparate data sets and to
in this way reach a higher coverage of information needs (Spruit and Vlug 2015).
It will not be necessary to integrate all disparate data sets; we showed that a very
good coverage of information needs can already be reached by integrating the three
most important data sets out of the total 14. One could set up so-called data spaces
which are loosely integrated sets of data sources where integration happens only
when needed (Hristidis et al. 2010). This could become an essential extension to
the earlier mentioned data exchange platforms so that these platforms offer—to a
certain degree—sensemaking of all the data sets that are shared through them. The
second avenue consists of tackling the lack of local and timely data. The Government
of Bangladesh has started to develop an online process of collecting the SOS and
D form data, tackling in this way the data granularity loss. We have co-created a
smartphone application in Bengali that local disaster management professionals and
volunteers can use to collect data just before and during the floods that fulfills the
currently not covered information needs. The functions and features of the app and
dashboard reflect the different clusters of information needs that we identified. The
data collected is fed back to the affected communities through a dashboard that is