Industry 4.0 and the reaction of the online public
The following looks into the perception of Malaysian Netizens (Specifically Twitter Users) on the topic of Industry 4.0 and check out the implications of Industry 4.0 on the Malaysian economy
The adoption of Industry 4.0, the digitization of manufacturing, has become a global initiative, with countries like Germany, India, and Southeast Asian nations, including Malaysia, pushing for digitalized industrialization. Malaysia has been proactive in implementing Industry 4.0, with its government's Industry4WRD initiative aimed at promoting the manufacturing sector's contribution to the economy. This paper attempts to conduct sentiment analysis by mining social media posts to understand the internet population's attitude towards Malaysia's Industry 4.0 initiative and discuss possible implications and results.
The paper follows a sentiment analysis approach to scrape data across several years on Industry 4.0 for malaysia using Machine Learning methods
and natural language processing in Python.
According to IBM, Industry 4.0 is a term used to describe the fourth industrial revolution that is currently underway in manufacturing and other industries. It involves the integration of new technologies such as Internet of Things (IoT), cloud computing, analytics, and AI and machine learning into production facilities and throughout operations.
It is important to note that the concept of 'Industry 4.0' is not a scientific concept and is something which is heavily debated upon by people in academia and also in various fields of Business and industry
The Consolidated Strategy on the Fourth Industrial Revolution for ASEAN highlights three areas of focus for Industry 4.0 in Southeast Asia: Technological Governance and Cybersecurity, Digital economy, and Digital Transformation of society. Singapore, Thailand, Indonesia, and Malaysia have launched initiatives related to Industry 4.0.
Malaysia announced its Industry 4.0 initiative in 2018, with a focus on enhancing the manufacturing sector, increasing labour productivity, promoting innovation capacity, and creating high-skilled jobs. The country aims to become a strategic partner for smart manufacturing in the Asia Pacific and attract more high-tech investments. Industry 4.0 in Malaysia also aims to fulfill the sustainable development goals set by the United Nations. The Malaysian government is enthusiastic about the initiative and plans to leverage digital technologies to stimulate industrial productivity in the region.
Based on the literature raised above, the paper asks the following research question:
The paper tries to answer the research question through the method of sentiment analysis using Natural language processing methods in Python to process the public sentiment of Malaysians towards the concept of 'Industry 4.0'
The paper discusses sentiment analysis in Python programming language as a way to identify the emotional tone behind a body of text. It helps organizations categorize opinions about their products, services, or ideas using data mining, machine learning, and artificial intelligence. Sentiment analysis is used in product analysis, marketing research, and social media monitoring to understand the social sentiment of their brand or product. The algorithm identifies, extracts, and quantifies the emotional tone behind a body of text.
For a dataset to perform Sentiment analysis, the paper mines/scrapes data off of Twitter with a search query which states 'Industry 4.0 in Malaysia'. The Code is as follows:
import snscrape.modules.twitter as sntwitter
import pandas as pd
query = "Malaysia Industry 4.0 until:2022-06-01 since:2015-01-01"
tweets = []
limit = 5000
for tweet in sntwitter.TwitterSearchScraper(query).get_items():
#print(vars(tweet))
#break
if len(tweets) == limit:
break
else:
tweets.append([tweet.date, tweet.user.username, tweet.content])
df = pd.DataFrame(tweets, columns=['Date', 'User', 'Tweet'])
df.to_csv('Malaysia Tweets.csv')
The code raised above will give us tweets dataset from 2022 to 2015 in relation to 'Malaysia Industry 4.0'
The dataset will look something like this:
0 2022-05-30 09:11:42+00:00 SkillstoProsper In Malaysia, S4P improves the quality, equity and relevance of skills and technical and vocational education and training (TVET) systems to improve youth employability prospects in the era of industry 4.0, through inclusive industry-led training.
1 2022-05-05 02:05:48+00:00 OfficialMIDA In light of the speedy growth of Industry 4.0 technologies in the region, Malaysia’s logistics industry is well-positioned to capitalize on the evolving digital landscape. https://t.co/sJjHzXvxlm
2 2022-04-25 06:48:04+00:00 InvestKL #Malaysia is gaining repute as a leading Asia-Pacific country in its readiness to adopt Industry 4.0 #technology, with a strong potential to adopt #robotics #automation. Some world-leading #MNCs have established their regional operations centre in Greater #KualaLumpur. https://t.co/UZ4om4Qx3R
3 2022-03-30 01:08:18+00:00 DhlExpressMY JETRO is 🤝collaborating with government agencies and Japanese companies to ☝️accelerate adoption of Japanese Industry 4.0 🏭 #smartmanufacturing technologies among Malaysian SMEs, as part of the Look East Policy (LEP) cooperation. https://t.co/3M9mtHUymT #malaysia #smemalaysia
4 2022-03-14 13:25:12+00:00 Dr_Nur_Mazlini Infographic: Is Malaysia Ready for Industry 4.0? | Manufacturing - Solidiance https://t.co/KksxvCJWDP via @
5 2022-03-09 08:58:44+00:00 MITIMalaysia It complements existing and upcoming national blueprints including the 12th Malaysia Plan, Shared Prosperity Vision 2030 Plan, National Policy on Industry 4.0 (Industry4WRD), as well as the upcoming New Industrial Masterplan. @AzminAli @limbanhong @lokmanhakim_ali #NIA #NIP
The next step is to conduct Sentiment Analysis with the dataset
Sentiment analysis is done by using a pre-trained model, which is loaded using Transformers package in python and then used to create sentiment scores for the whole dataset
The code to do sentiment analysis is as follows:
from cgitb import text
import csv
import pandas as pd
import nltk
import numpy as np
from transformers import AutoTokenizer
from transformers import AutoModelForSequenceClassification
from scipy.special import softmax
from tqdm.notebook import tqdm
from nltk.sentiment import SentimentIntensityAnalyzer
df = pd.read_csv("Malaysia Tweets.csv")
example = df['Tweet'][50]
MODEL = f"cardiffnlp/twitter-roberta-base-sentiment"
tokenizer = AutoTokenizer.from_pretrained(MODEL)
model = AutoModelForSequenceClassification.from_pretrained(MODEL)
#Running example on ROBERTA
encoded_text = tokenizer(example, return_tensors='pt')
output = model(**encoded_text)
scores = output[0][0].detach().numpy()
scores = softmax(scores)
scores_dict = {
'Roberta_Neg' : scores[0],
'Roberta_Neu' : scores[1],
'Roberta_Pos' : scores[2]
}
def polarity_scores_roberta(example):
encoded_text = tokenizer(example, return_tensors='pt')
output = model(**encoded_text)
scores = output[0][0].detach().numpy()
scores = softmax(scores)
scores_dict = {
'roberta_neg' : scores[0],
'roberta_neu' : scores[1],
'roberta_pos' : scores[2]
}
return scores_dict
#Whole Data set on ROBERTA
res = {}
for i, row in tqdm(df.iterrows(), total=len(df)):
text = row['Tweet']
myid = row['User']
#roberta_result = polarity_scores_roberta(text)
res[myid] = polarity_scores_roberta(text)
print(pd.DataFrame(res).T)
pd.DataFrame(res).T.to_csv('Malaysia Sentiment ROBERTA.csv')
The overall result will gives us the sentiment scores of all the tweets in the dataset and then will give us the following result:
| Figure 2 Overall Sentiment Score for all the tweets collected on Malaysia Industry 4.0 | |||
| 'roberta neg' Indicates the negative sentiment probability of the tweets | 'roberta neu Indicates the neutral sentiment probability of the tweets | 'roberta_pos' Indicates the positive sentiment of the tweets | |
| The overall sentiment of all the tweets | 0.074194 | 0.607835 | 0.317971 |
| 7.42% | 60.78% | 31.79% | |
| Total number of tweets: 702 | |||
The results convey that most of the tweets in the dataset are Neutral Positive towards the concept of 'Industry 4.0 in Malaysia'
The results conclude that most of the tweets and their contents towards Malaysia Industry 4.0 have been neutral to positive. The negative probability of all the tweets is as low as 7.42%, which indicates that most Twitter accounts do not think of this Industry 4.0 in Malaysia as a flawed initiative. The total probability score of Neutral-Positive sentiment is about 92.57%, which is a very optimistic score for the initiative in Malaysia.
The paper’s analysis infers that most of the public’s sentiment towards the initiative of Industry 4.0 in Malaysia is likely Neutral-positive on the social media platform Twitter. The result infers that on social media, at least twitter in general, users from around the world look at Malaysia’s Industry 4.0 attempt with a relatively good optimism and the public opinion might not necessarily create any hindrance in the enactment of the policy.