Abstract
The Fourth Industrial Revolution (4IR), driven by digital technology
is transforming consumer preferences and provides
opportunities for novel digitalisation networking tools to attract
and connect with consumers.
Introduction
The significance of social media in customer service cannot be overstated,
particularly as the 4IR progresses and the desire for automated
customer service approaches in customer-centric environments increases.
Internet and various social media platforms are the preferred
methods of communication for consumers in recent times. The advent
of 4IR technologies has paved the way for customer engagement,
which is unavoidable for Small, Medium and Micro Enterprises
(SMMEs) that use social media for marketing and engagement purposes.
Artificial intelligence (AI) systems are capable of making some
decisions without human involvement. To advance the digital business
operation using AI technology, we present a tool for the automatic
generation of responses to social media comments.
Problem statement
Responding to enormous comments on social media platforms is
one of the biggest challenges facing businesses in recent times, especially
when dealing with irate consumers. Customers have increasingly
adopted social networks as a platform for expressing their concerns
and posting comments on business pages, posing a great challenge
for customer support agents and digital marketers alike. Analysing
and responding manually to these enormous comments is a timeconsuming
task, necessitating the adoption of a Natural Language
Processing (NLP) algorithm that can complete the task swiftly — automatic
comprehension of social media posts for comment generation.
Methodology
This dissertation presents AI algorithms and a tool for the automatic
comprehension of customer tweets and the generation of responses to
these tweets. This artefact was designed using the Design Science Research
(DSR) approach, which takes a structured approach to solving
a challenge. This was done in two-fold: using existing NLP libraries
to preprocess and tokenise these tweets, and using rule-based algorithms
to find a matching response to each customer based on the array
of extracted tokens from the customer’s tweet. This was built into
a tool called Comment-Synthesizer. This tool takes unfiltered tweets
as input, preprocesses the tweets, and matches the tweets with previi
defined responses using a rule-based algorithm. If implemented in a
desktop automation application, this tool can respond automatically
to a massive number of comments or posts on social media from customers.
Results
The design of the tool, along with the results corresponding to the
concepts were presented. The tool was implemented and built into a
desktop application. This application shows a simple interface that allows
users to type in their tweets and receive an automated response
based on that tweet.
Conclusion
The 4IR technological artefact presented in this dissertation has the
potential to impact customer-centric environments. This artefact was
designed using the DSR methodology. We demonstrate the deployment
of the tool with the results of the responses it generated. The
evaluation of the results suggests that this technology will be helpful
in the advancement of SMMEs in South Africa.