Watson is essentially a platform by IBM that is based off of the
fundamentals of machine learning algorithms to provide several
different analytical services amongst other services to clients. Using
the variety of services offered by Watson, clients are able to
transform business processes from data perspective mostly.
There are several different algorithms used in the platform, however,
the main component to Watson's machine learning is Natural
Language Processing (NLP). NLP is an artificial intelligence branch that
provides the ability to understand spoken word or text in the same
way we can. It combines different aspects of computational linguistics
such as rule-based models of the human language with statistical, ML
and DL techniques. Watson heavily relies on NLP to automate
complex business processes while gaining insights on that data. There
are several different tools under the Watson umbrella that employ
NLP derived tools for this.
In my professional experience, I have seen a lot of data go unused due
to the fact that it is mostly hard to catalog along with organizational
structure issues that cause silos within one agency. I am a financial
consultant for an agency working with budget formulation teams
within each department of the agency. While most of the agency's
issue is a question of organizational structure, the data within the
agency is somewhat consistent yet uncatalogued. Since the agency
has regulations and standards for budgetary data, it would be fairly
simple to organize all the data under one catalog. I found that the
Watson Knowledge Catalog could be a useful tool in organizing the
data for the agency's accounting purposes. Right now, the agency's
budgetary data is segregated by department and in simple data
warehouses that do not employ any ML/AI whatsoever. We currently
only use straight-line projections along with some non-ML factoring
and qualitative measures into our budget formulation. As someone
with insight into the metadata as well as the required analysis for
forecasting, I see a lot of missed opportunities for implementing better
data analytical models. The Watson Knowledge Catalog also caught
my eye due to its ability to control access to ensure appropriate use.
This is very important as the departments in the agency essentially
compete over the budget allocated to the agency and as a result are
highly protective of their budgetary data. This aspect of the tool
would allow the appropriate permissions to be set even if all the data
is under one roof.
Another tool I found useful under the Watson catalog is the Statum
KPI. This is a tool that helps banks directly measure and manage their
performance. I thought this performance management solution could
be restructured in a way to be applied to the different departments
within the agency I work for. This tool would allow the agency to
measure and manage their performance between the departments
within the agency. This is important when looking at the department's
spending in comparison to their allocation. Around the budget
formulation period, the departments would request a certain amount
to be allocated for their budget. This request is compared with their
spending and previous year allocation. However, with the use of this
tool, we can breakdown their efficiency and understand in which
aspects they are performing well and which aspects did not yield
expected results. With this insight, we can appropriately allocate the
budget to the specific areas based on the KPI indicators discovered
through this Watson tool.
P.S.
If anyone would like to kindly provide data for my chatbot, it would be
greatly appreciated. Thank you!
https://web-
chat.global.assistant.watson.cloud.ibm.com/preview.html?region=us-
south&integrationID=3d4c4e6d-7410-4c4a-9728-
3092f1611e95&serviceInstanceID=999f902b-87db-41fa-98f4-8e7fed4f4d8f
