Analysis on the ABC NEWS Report on Huawei Based on Data Analysis
Zhecheng Xiao
MUCCI, Monash University, Melbourne, 3161, Australia
Keywords: Huawei, News Media, Data Visualization, Cluster, International Relations, Data Analysis.
Abstract: With global politics development, the confrontation between China and the United States will become the
norm. As the most famous Chinese communication company, Huawei often becomes a victim of ideology.
Huawei has been deeply involved in communications construction in Australia in the past decade. Therefore,
as a traditional ally of the USA and UK, the Australian government's attitude towards Huawei is also an
important aspect that can be studied. The different tendencies of media reports can reflect the international
situation to a certain extent. Simultaneously, media act as the government's mouthpiece. Exploring the
emotional changes in media can reflect the government's attitude to a particular time. This paper mainly
studies how Australian media reports news related to Huawei. ABC News, one of Australia’s largest media,
was chosen as objectives, as the research object because its coverage is more comprehensive than other
Australia media. Based on data analysis, the emotions contained by these news were examined expressed in
these reports and produced word clouds. This paper selected 943 news about Huawei from ABC NEWS as
its research objects to achieve the above goal. Python language was used to crawl. Through analyzing ABC's
news data about their titles and contents, it is found that there is a positive correlation between the number of
reports and events. At the same time, most of the reports on Huawei in 2019 and 2020 have a negative
tendency.
1 INTRODUCTION
The U.S. government began sanctioning Huawei
companies for the first time in January 2018. The U.S.
government was firmly opposed to Huawei signing
cooperation with U.S. telecom operator-AT&T and
banning Huawei mobile phones from entering the
U.S. market. Over the next few months, U.S.
lawmakers continued to claim that Huawei phones
had "back doors" and stolen user data, calling on allies
to take the same measures to limit Huawei phones. As
a result, in July of the same year, Australia took the
lead in banning Huawei from participating in the 5G
construction on national security grounds. Although
Australia always supports the U.S., the Australian
government's attitude towards Huawei has been
wavering.
Huawei established its first local board of directors
in Australia as early as 2011 as a pilot of its
globalization strategy (Cai, 2015). Australia's council
of directors' most important work is lobbying the
government and building good relations with the local
media community (Kania, 2018). Huawei even
bought an Australian football team to increase its
influence in the Australian market. Huawei wanted to
step into Australia's national broadband program in
2012, although the Australian government finally
rejected it on national security grounds; Huawei also
participated in constructing and 5G network
construction. Huawei's success in the mobile phone
market has triggered a strong rebound in the U.S.
government and raised Australia’s concerns. The
natural differences in the political positions of Huawei
companies and the Chinese government with Europe
and the United States lead to the Australian
government distrusting. The Australian government
believes that the Huawei network lacks transparency
and may be used by the Chinese government
(Jennings, 2018).
Under these social conditions, a study of Huawei
reporting changes in Australian media is of great
value. The author takes ABC news as the research
object, hoping to explore its reaction to the rejection
of Huawei by the Australian government. In this
article, the author will use news data to analyze how
the Australian local media reported significant
changes in multinational companies. For example,
how the attitude of the same news media changes in
the course of events, how the importance of
120
Xiao, Z.
Analysis on the ABC NEWS Report on Huawei Based on Data Analysis.
DOI: 10.5220/0011731300003607
In Proceedings of the 1st International Conference on Public Management, Digital Economy and Internet Technology (ICPDI 2022), pages 120-125
ISBN: 978-989-758-620-0
Copyright
c
2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
information dissemination has changed, how the
content direction of information dissemination
changes. Finding experiences and lessons from this
news communication’s characteristics and improving
future cross-cultural business activities is also one of
the project’s problems.
2 FARMING THEORY
The frame theory of news is different from the
traditional news value and goalkeeper theory. This
theory can reflect the value judgment of reporters
when choosing news and events (Brüggemann, 2014).
The frame structure constitutes the fundamental
thinking and presentation of NEWS. Therefore, by
analyzing news reports’ framework, the author can
understand the connotation of reports more
intuitively. The news reports on the website are
mainly composed of a title, author, release time, news
description, news text, keywords and pictures. The
unified structure standard of news website constitutes
the basic news framework and is also the basis of this
analysis of news data.
3 RESEARCH METHOD
3.1 Data Capture and Cleaning
The author chose to use python to grab news because
it is simpler and more operable than C++. For
example, data statistics and a certain degree of
visualization can be implemented. However, in the
actual operation process, the author found that only
python cannot directly grab ABC news sites.
Therefore, the author used the scape framework to
help crack the site's anti-grabbing protection. General
processes are shown in figure 1 (Fan, 2018).
The author used Request Object to get all the
information of the whole page, including text,
pictures, Cookie, client certificates, query strings, etc.
Therefore, the web pages’ layout information
obtained from news pages contains a large amount of
data that people do not need to use when analyzing
problems, such as the code in the header file and the
code related to the web pages’ layout. Using Beautiful
soap can help extract the data we want from the
HTML files of a web page, such as URL, news
headlines, news organizations.
The author got a table about Huawei report in ABC
NEWS through the above data capture and cleaning
method. At first, the author wanted to choose 2012
Huawei entering the Australian market as the starting
point of the data. However, the author found that the
sample size was too small to be representative.
Finally, the author chose 2018, Huawei was banned
from Australia's national broadband construction as
the starting year and captured the data was until 2021.
The lines of the table include the news title, release
time, primary content. Finally, the total amount of data
the author got is 1062; through manual screening, the
final valid data’ number is 943.
Figure 1: Data capture process.
Analysis on the ABC NEWS Report on Huawei Based on Data Analysis
121
3.2 Keyword Trend
The consistent structure of a web page makes crawling
data possible. After grasping the keywords, the author
can start analyzing the keywords. First, we can use the
Term Frequency -Inverse Document Frequency (TF-
IDF) method for keyword analysis. Some words are
classified as keywords because they appear more
frequently (Roelleke, 2013). Quantitative data is the
purpose of TF-IDF calculation. After quantifying the
data, the content of the data can be classified by
keywords. News keywords can introduce emotional
classification and category tagging. The author can
then find whether the emotion of news is positive,
neutral or negative by keyword clustering (Topkev,
2016). To judge the tendency of the report. In addition
to using data to express news analysis results of news
analysis, researcher can also use pictures, tables and
colors to express.
On the choice of tools for data analysis and
visualization, the author chose Python and Dychart.
Dychart is a software that can easily visualize data,
and the built-in preset template can greatly reduce the
time of making visual charts. Although the author has
captured the relevant headlines of the news in the data
collection phase, and direct analysis of the title is the
fastest and easiest way, the analysis of news headlines
may not fully reflect the changes in Huawei by news
reports. Hence, in the end, the author chose news
content as the source of keyword capture.
After selecting the text as the source of the
keyword, the author need to face the most intuitive
problem: there are many recurring words. There is not
much practical significance and does not play a
decisive role in our research, such as punctuation and
prepositions or conjunctions. Hence, the author use
the python data cleaning function (data clean) to filter
punctuation. At the same time, English words contain
many changes in tense and plural. At this point, we
need to use morphological repair. Lexicalization can
transform words into general forms and express
complete semantics. For the first time, this will clean
up the keywords and then manually screen out
unnecessary conjunctions such as “and”, “or”, “then”.
Add these words to the stop list and filter the
keywords again to select the first 100 words with the
highest frequency. Subsequent analysis will also be
based on chosen keywords.
3.3 Word Cloud Map
Word cloud analysis is also the primary method of
visualizing news data. Cloud images can convey
information about the importance of words through
the size of terms. The word cloud image can also be
distinguished by adding different colors to different
words. Simultaneously, the word cloud map can
intuitively show the important difference of keyword
and display the key information of news text
(Ponnambalam, 2019). By forming a word cloud, the
author can filter out most meaningless textual
information, and words with specific meanings can be
retained. However, prepositions or meaningless
function words still need to be cleared in advance.
The author chose to make word clouds to visualize
keyword changes in 4 years. The word cloud can
visually display keywords in news reports by the size
and color of the text. Import the obtained keyword and
word frequency data into Dychart. Using the word
cloud template, simply adjust the text size and color to
generate the corresponding word cloud picture. The
word clouds of 2018-2021 to express ABC News
reports’ keywords can be got.
3.4 Text Clustering Analysis
Based on keywords and TF-IDF analysis, the news
can be clustered to discuss the report’s emotion.
Hence, KMeans was used to classify the text. The
author imported the KMeans Library in Python. In the
final output, yellow represents that the article has
positive emotions, and black represents that the article
has negative feelings.
4 RESULTS
The researcher finds that as shown in figure 2. In July
2018, the Australian government took the lead in
banning Huawei from participating in Australia's 5G
network on national security grounds, and the number
of news reports showed a rapid growth trend. In mid-
2019, with the release of the Trump administration's
Defense Authorization Act, the number of reports
peaked at 96 and 94 in a month, which has been called
a high concern for single media. In 2020, the
Australian subsidiary of Huawei announced the
termination of its ten-year sponsorship relationship
with the Australian rugby team Canberra Raiders,
which was also continued reported on ABC news for
a while (Overton, 2020). Thus, ABC News reports on
Huawei are positively related to the Australian and
American governments’ continued attention.
ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology
122
Figure 2: The number of reports in 2018-2021.
Not difficult to see from the figure 3, 4, 5 and 6.
Even ABC News is an Australian media, China and
the US are still some of the most frequent words. This
shows that the Australian market is in a sense a vane
of American policy. The 2018 cloud chart shows how
the United States and its allies, such as Canada,
impose sanctions on Huawei. In 2020, new Chinese
software such as TikTok can be seen joining the story.
It is mainly around Huawei and the US Trump
government how to play the corresponding game in
2019. It can be seen that 2019 is also the year ABC
News reports the most significant number of Huawei.
Figure 3: 2018 Word Cloud Map.
Figure 4: 2019 Word Cloud Map.
Figure 5: 2020 Word Cloud Map.
Figure 6: 2021 Word Cloud Map.
Based on this set of text clouds, ABC news
attention’s focus can be found from the initial report
on the situation between Huawei and the world to the
game between China and the United States. There was
also a concern in 2021 about the impact of the U.S.
election on Huawei companies and whether the Biden
government and Trump's attitude to Huawei will
change. As the United States sanctions and
development on Huawei and other Chinese
technology enterprises, the news media’s focus is
shifting to track the progress of the game between
China and the United States from the perspective of
media attitudes and keywords.
From figure 7, it can be seen the position and
emotional tendency of ABC News on Huawei. Some
Analysis on the ABC NEWS Report on Huawei Based on Data Analysis
123
Figure 7: Cluster analysis.
of the results are shown in the figure above.
Researcher find that ABC News reports on Huawei
are positive most of the time. However, there will still
be a lot of adverse sentiment reports in 2019 and 2020.
The different changes may be caused by the Sino US
trade war and the Australian government’s dialogue
measures.
5 CONCLUSION
This research finds that news reports’ number and
tendency are related to the international situation.
There are related essential events at the peak of each
report.
In the process of data analysis, there are still many
areas that need to be improved. For example, although
the author imported a stop word, the exported
keyword still contained some meaningless words.
This is because the stop words were manually filtered,
and there were some omissions in the selection. In
addition, the code when writing programs can be
optimized. If future research can calculate the
similarity between different documents, the study will
become more complete.
REFERENCE
A. Topkev. Discourse analysis and media attitudes: the
representation of Islam in the British press [Review of
Discourse analysis and media attitudes: the
representation of Islam in the British press]. Journal of
Multilingual and Multicultural Development, 2016,
37(2), 216–218. Routledge. https://doi.org/10.1080/
01434632.2014.973654
E. Kania, & Australian Strategic Policy Institute. Huawei
and Australia’s 5G Network (pp. 04-05, Rep.).
Australian Strategic Policy Institute. 2018. doi:10.
2307/resrep23079.
K. Ponnambalam, & Linkedin.com. Text Analytics and
Predictions with Python Essential Training. 2019.
M. Brüggemann. Between Frame Setting and Frame
Sending: How Journalists Contribute to News Frames.
Communication Theory, 2014, 24(1), 61–82. https://
doi.org/10.1111/comt.12027
P. Jennings, & Australian Strategic Policy Institute.
Huawei and Australia’s 5G Network (pp. 25-26, Rep.).
Australian Strategic Policy Institute. 2018. doi:
10.2307/resrep23079.15
P. Overton. Huawei quits Raiders: Telco giant Huawei has
blamed the trade dispute between Australia and China
for its decision to cancel its major sponsorship deal with
the Canberra Raiders. In Nine News. 2020.
ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology
124
T. Roelleke, & Morgan & Claypool Publishers.
Information retrieval models foundations and
relationships. Morgan & Claypool. 2013.
Y. Fan. Design and Implementation of Distributed Crawler
System Based on Scrapy. IOP Conference Series. Earth
and Environmental Science, 2018, 108(4), 42086.
https://doi.org/10.1088/1755-1315/108/4/042086
Y. Cai. How does Huawei "go out". 21st Century Business
Review (1), 2015, 30-31.
Analysis on the ABC NEWS Report on Huawei Based on Data Analysis
125
