Analyzing the discussion of gregorio murder on twitter using text mining approach

Abstract

Being dubbed as the “social media capital of the world”, the impact of social media among Filipinos went beyond information dissemination. Filipinos also used social media to call out authorities and revolutionize political systems. Using text mining approach, this study explored the common themes surrounding the social media users' discussions regarding the recent killing of Sonya and Frank Gregorio which sparked conversation on Twitter in the Philippines. Using keywords from the trending hashtags #StopKillingsPH and #JusticeforSonyaGregorio, tweets were extracted via Python program. A total of 1045 tweets from December 21 to December 28, 2020 were collected and analyzed in terms of frequency, sentiment, subjectivity, and surrounding themes. Results show that discussions regarding the Gregorio murder revolved on the netizen's call for justice, demand that the suspect be charged with murder, and an online petition and a louder plea to stop killings and end police brutality in the country. Results of this study are discussed in evidence on how social media can be influential in mobilizing the authorities to act against injustices in the Philippines and in affecting change in the society.

Full text

Computers in Human Behavior Reports 8 (2022) 100248
Available online 15 November 2022
2451-9588/© 2022 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Analyzing the discussion of gregorio murder on twitter using text
mining approach
Merimee Tampus- Siena
a
,
b
,
*
a
Psychology Department, Ateneo De Manila University, Philippines
b
Ofce of Student Affairs and Student Services, Philippine Normal University, Philippines
ARTICLE INFO
Keywords:
Text mining
Twitter
Social media
Philippines
ABSTRACT
Being dubbed as the “social media capital of the world”, the impact of social media among Filipinos went beyond
information dissemination. Filipinos also used social media to call out authorities and revolutionize political
systems. Using text mining approach, this study explored the common themes surrounding the social media
users’ discussions regarding the recent killing of Sonya and Frank Gregorio which sparked conversation on
Twitter in the Philippines. Using keywords from the trending hashtags #StopKillingsPH and #JusticeforSo-
nyaGregorio, tweets were extracted via Python program. A total of 1045 tweets from December 21 to December
28, 2020 were collected and analyzed in terms of frequency, sentiment, subjectivity, and surrounding themes.
Results show that discussions regarding the Gregorio murder revolved on the netizen’s call for justice, demand
that the suspect be charged with murder, and an online petition and a louder plea to stop killings and end police
brutality in the country. Results of this study are discussed in evidence on how social media can be inuential in
mobilizing the authorities to act against injustices in the Philippines and in affecting change in the society.
1. Introduction
Philippines is a country that is very accustomed to social media (Lee
et al., 2013), making the country dubbed as the “social media capital of
the world” for reigning in terms of social media use globally (ABS-CBN
News, 2019) and ranked 11th among Twitter use worldwide (Statista,
2021). Social media users in the Philippines do not only use social media
to connect with families and friends online, it is also utilized to learn
about recent news and happenings real time and disseminate
second-hand information (Takahashi et al., 2015).
Current news, events, and issues can spur people to talk about their
opinions, feelings, and experiences in social media. To put context or
information into an online post or tweet, social media users use “hash-
tags” such as #BlackLivesMatter or #COVIDpandemic (Lee et al., 2013).
They can subscribe and get updates by following these hashtags. To even
engage themselves with the topic, they can re-tweet another person’s
tweet or post their own opinion of the topic using the same hashtag. This
re-tweet option spreads the message beyond the recipients of the orig-
inal tweets (Abbasi et al., 2018). Twitter ranks trending tweets based on
the frequency each hashtag was used. One’s political behavior is shaped
by how they interact with their digital environment, including how they
behave in social media (Gainous et al., 2020). Social and political
activism in social media pose a great potential to reshape modern de-
mocracy as a public political instrument (Maulana, 2017). Following
this line, this study adopts the Social Media Engagement Theory which
posits that social media engagement, as a context-specic phenomenon,
constitutes cognitive, affective, and behavioral dimensions (Dessart,
2017). The information people get from social media help build their
political attitude and these resulting attitude shape their political
behavior (Gainous et al., 2020).
One prominent instance when Filipinos expressed their strong sen-
timents online was the death of the Gregorios. In December 20, 2020,
the identied suspect Police Senior Master Sergeant Jonel Nuezca shot
Sonya Gregorio and her son Frank Gregorio to death (Tantuco, 2020).
News reports said that the killing happened because of a confrontation
between the suspect and the victims over the ring of a “boga”, an
improvised noise-maker (Gonzales, 2020; Ranada, 2020). Nuezca went
to the residence of the victims to investigate who was shooting the boga
and tried to arrest Frank and his mother, Sonya Gregorio intervened.
Further, the suspect felt insulted after his daughter shouted to the vic-
tims that her father is a policeman and the female elderly victim
answered she does not care. This triggered the suspect to eventually pull
* Psychology Department, Ateneo De Manila University, Philippines.
E-mail address: merimee.tampus-siena@obf.ateneo.edu.
Contents lists available at ScienceDirect
Computers in Human Behavior Reports
journal homepage: www.sciencedirect.com/journal/computers-in-human-behavior-reports
https://doi.org/10.1016/j.chbr.2022.100248
Received 24 January 2021; Received in revised form 10 November 2022; Accepted 14 November 2022

Computers in Human Behavior Reports 8 (2022) 100248
2
his trigger and shoot both victims dead. Police reports revealed that the
suspect and the victims have long had disputes over right of way.
The video of the killing was posted in social media by one of the
relatives of the victims and spread like wildre which angered social
media users. The netizens used #StopKillingsPH, #PulisAngTerrorista,
#EndPoliceBrutality, and #JusticeforSonyaGregorio in expressing their
opinion about the incident online. In December 21, the #StopKillingsPH
and #JusticeForSonyaGregorio ranked rst and fourth respectively on
Twitter (Dominguez- Cargullo, 2020) following the video of police of-
cer Jonel Nuezca shooting dead Sonya Gregorio and her son Frank
Gregorio going viral on social media. The incident, even when reiterated
as an isolated case by the Philippine National Police, was connected by
social media users with the numerous killings committed by police of-
cers in 2020 alone, thus the call to stop these killings in the country
using #StopKillingsPH. From these cases, justice was not served well for
the victims given that the offenders were authorities themselves. This
urged social media users to use #JusticeForSonyaGregorio to demand
for justice for the murder of the Gregorios.
With these trends in the use and relevance of social media during
controversies, issues, and problems in the society, it is important to use it
as a tool to analyze the attitudes and opinions of social media users
which cannot be gathered using traditional methods (like interviews and
surveys). There are many studies in the Philippines which examined
online engagement of Filipinos, particularly the usage of social media
during disasters (Cabotaje & Alampay, 2013; Lee et al., 2013; Takahashi
et al., 2015). These studies showed how social media has been instru-
mental in helping the government respond to natural disasters. Another
study by Umali, Miranda and Ferrer in June 2020 looked into the sen-
timents of Filipinos on the selected government agencies in the country.
Meanwhile, there is a dearth of studies which looked into the anal-
ysis of discussion on Twitter involving a more controversial topic, un-
lawful killings in particular. This study aimed to address this gap by
looking into how social media created an impact in revolutionizing
political systems and mobilizing authorities to respond and bring justice
to unlawful killings in the Philippines. This study used the trending
hashtags #StopKillingsPH and #JusticeforSonyaGregorio (Domi-
nguez-Cargullo, 2020) to extract tweets in the people’s stances towards
the given topic on Twitter in order to explore the common themes sur-
rounding the Twitter users’ discussions regarding the recent killings of
the Gregorios. This study aims to shed light on the relevance of social
media in the social issues in the Philippines, particularly in analyzing its
usage and relevance in mobilizing authorities to nd justice for victims
of unlawful killings.
2. Methodology
To achieve the objectives of the study, the researcher started by
collecting publicly available social media data from Twitter using the
Python program. Different keywords were used to extract the tweets
such as “Stop Killings PH” and “Justice Sonya Gregorio”. The methods
followed by this study is presented in Fig. 1.
The accessed tweets were combined in one spreadsheet. Sample
tweets include “Because we really need/want the killings in the PH to
stop” and “Even the young presidential apo (grandson) asked to put an
end to the unlawful killings.” Table 1 shows the initial tweets extracted.
The tweets were then cleaned and pre-processed to avoid characters
with no meaning for the scope of the analysis and to easily extract in-
sights from the data set. Stop words (i.e. articles, pronouns, punctua-
tions, white space) were deleted. Table 2 shows the cleaned tweets.
Data was organized in terms of frequencies from the Term Frequency
Inverse Document Frequency (TFIDF). TFIDF is an algorithm which
considers the frequency of the word and in how many times a term can
be found in a le (Hakim, Erwin, Eng, Galinium, & Muliady, 2016). It
calculated the weight of each word and reported only the relevant words
appearing in the entire corpus. TFIDF was used because it ascribes
importance to the relevant words appearing in the entire corpus by
considering every weight of the word. It adjusted the frequency of terms
by their rarity in the document (Christian et al., 2016) and when a word
was found in so many les, it will be considered by the program as
unimportant (Hakim et al., 2016). For example, the term with the
highest frequency from the collected tweets was actually the word
“justice” but when TFIDF was used to calculate the weight of each term
in the lexicon, “murder” now became the most frequent term. This is
because the terms “justice” and “murder” have two different weights and
since “justice” appeared too frequently in the corpus, it was considered
unimportant or of less weight. TFIDF was used in this paper in order to
extract the most relevant terms among the tweets and disregard the less
important terms.
Sentiments and subjectivities were also analyzed in order to learn
about the associated emotions in the collected tweets. Cloud of words
was also generated to get a general overview of the collected data.
Lastly, Latent Dirichlet Allocation (LDA) was used to classify the text in
the document to a particular topic; thus, identifying the themes sur-
rounding the tweets.
Fig. 1. Methods followed in the study: data collection, data pre-processing, and
data analysis.
Table 1
Initial tweets extracted through text mining.
created_at text
December 28, 2020
10:19
not jam magno calling kpop stans that are lipino youths
bobo and uneducated for trending the hashtags to stop polˆ
a
€
¦
https://t.co/7oPFNJDAjB
December 28, 2020
5:57
i think in general she saying that ˆ
a
€
œkpop stansˆ
a
€
•or people
who spread abt whats happening in the ph like theˆ
a
€
¦ http
s://t.co/MGzWXGaP1h
December 27, 2020
6:03
Because we really need/want the killings in the PH to stop.
EJK, War on Drugs, Redˆ
a
€
¦ https://t.co/IqkmKTSzEG
December 25, 2020
12:39
Even the young Presidential apo asked to put an end on the
unlawful killings. https://t.co/9KOwR2mcj0
December 24, 2020
17:48
The killings will never stop!!! https://t.co/WN7xsWgV3g
M. Tampus- Siena

Computers in Human Behavior Reports 8 (2022) 100248
3
3. Results and discussion
A total of 1045 tweets from December 21 to December 28, 2020 were
collected and analyzed from the tweets extracted from Twitter using the
keywords “Stop Killings PH” and “Justice Sonya Gregorio”. Most of the
tweets were posted on December 22 (80.28%).
3.1. Word analysis
The top 10 most frequently used words surrounding the tweets on the
discussion of Gregorio Murder on Twitter from the TFIDF matrix were
reected in Fig. 1. Among the extracted tweets, the Python program
generated only the ten most frequent words in the corpus.
The terms “murder”, “charge”, “ofcer”, “nuezca”, and “jonel” have
the highest frequency in terms of the number of occurrence in the
collected tweets. These tweets reect how the social media users wish to
charge the suspect, Jonel Nuezca with murder for killing the Gregorios
(see Fig. 2).
Helping the visualization, a cloud of words was generated and is
presented in Fig. 3. Words such as “sonya”, “frank”, “gregorio”, “de-
mand”, “justice”, “jonel”, “nuezca”, “charge” reect that the discussion
of Gregorio murder on Twitter demands for justice for the death of
Sonya and Frank Gregorio. This justice can be achieved when the suspect
Jonel Nuezca is charged of murder.
3.2. Sentiment and subjectivity analysis
Sentiment and subjectivity analysis helped in interpreting the tweets
of the social media users that are based on the polarity (positive and
negative) of their own personal opinion (Umali et al., 2020) through the
use of lexicons.
Fig. 4 shows the visualized distribution of sentiment and subjectivity.
The sentiment of the tweet depends on the polarity of each word found
in a tweet. A score of +1 is given to a positive word, for example best,
good; a score of −1 is given to a negative word, for example bad, worse;
and 0 is given for neutral words, for example average, quite (Nausheen
& Begum, 2018). On the other hand, subjectivity, of which score ranges
from 0 to 1, reects the views of the twitter users towards the murder of
the Gregorios where the score of 0 means that the tweet has an objective
content while a score of 1 means that the tweet has a subjective content
(Nausheen & Begum, 2018). From these values, the gure shows that the
polarity of words between positive and negative in all the tweets are
Table 2
Cleaned and pre-processed tweets.
created_at text cleaned_text
December 27,
2020 6:03
Because we really need/want
the killings in the PH to stop.
EJK, War on Drugs, Redˆ
a
€
¦ http
s://t.co/IqkmKTSzEG
really needwant killings ph stop
ejk war drugs red
December 25,
2020 12:39
Even the young Presidential
apo asked to put an end on the
unlawful killings. https://t.
co/9KOwR2mcj0
even young presidential apo
asked put end unlawful killings
December 24,
2020 17:48
The killings will never stop!!!
https://t.co/WN7xsWgV3g
killings never stop
December 23,
2020 14:45
Yes, it wont. But it will deter
some, specially premeditated
murders. Killings wonˆ
a
€
™t
stop killings, Bishop Pabilloˆ
a
€
¦
https://t.co/Rcg8ISnq9d
yes wont deter specially
premeditated murderskillings
wont stop killings bishop pabillo
December 23,
2020 7:34
please understand that there
are ways to raise awareness for
ˆ
a
€
œstop the killings in PHˆ
a
€
¦
without bringing down blm.
iˆ
a
€
¦ https://t.co/jBgTXfwcy2
please understand ways raise
awareness stop killings ph
without bringing blm
Fig. 2. Frequency of relevant words in the tweets surrounding the discussion of Gregorio Murder.
Fig. 3. Word cloud generated from the tweets surrounding the discussion of
Gregorio Murder.
M. Tampus- Siena

Computers in Human Behavior Reports 8 (2022) 100248
4
seemingly balanced and neutral. The line indicates that these neutral
values are rather close to zero.
Looking into the results of the sentiment analysis, Fig. 5 shows that
the tweets surrounding the discussion of Gregorio Murder are more
neutral and only a few tweets were scored greater than or less than 0. For
example, a tweet saying “The killings will never stop!” was scored 0 or
regarded as neutral while the tweet saying “even the young presidential
apo (grandchild) asked to put an end to the unlawful killings” was scored
0.10, still regarded as neutral.
On the other hand, Fig. 6 shows that the result of the subjectivity
analysis of the tweets is mostly zero which mean that the words used in
the tweets surrounding the discussion of Gregorio Murder are more
objective than subjective. For example, the tweet “because we really
need/want the killings in the PH to stop, EJK (extrajudicial killings), war
on drugs” was scored 0.2 or regarded as rather objective.
The results of the sentiment and subjectivity analysis reect that the
tweets surrounding the murder of the Gregorios primarily called for
justice and demanded that these kinds of unlawful killings in the
Philippines to stop. The individuals who posted these tweets expressed
their feelings in a rather neutral way and expressed their views more
objectively.
3.3. Thematic analysis
In this study, themes were analyzed using an approach called topic
modeling which aims to extract main topics that occur in a corpus.
Latent Dirichlet Allocation (LDA) was used to classify the text in the
document to a particular topic. LDA is viewed as a dimensionality
reduction technique (Blei et al., 2003) in natural language processing. It
looks into why some parts of the data are similar (Blei et al., 2003) and
assumes that each corpus is a mixture of a small number of topics and the
presence of each word is attributable to one of these topics. Table 3
presents the themes identied based from the results of the analysis.
Theme 1: Demand justice for Sonya and Frank Gregorio by charging
ofcer Jonel Nuezca with murder. The discussion of the Gregorio murder
on Twitter was mainly about the social media users’ demand for justice
for what happened to the victims. For them, justice is served when the
suspect, police ofcer Jonel Nuezca is charged with murder and even-
tually put to jail. Nuezca eventually surrendered himself an hour later
after the incident (Ranada, 2020). He is now facing two counts of
murder and the government promised a thorough investigation of the
case (Reuters, 2020). Sonya and Frank Gregorio were laid to rest on
December 27.
Theme 2: Call up the Department of Justice through an online petition to
give justice for Sonya and Frank Gregorio. Twitter has been a platform for
the social media users to gather online petition to give justice for the
murder of the Gregorios. With this online petition, they wish to be heard
by the authorities, particularly the Department of Justice, who will
respond to the case properly and serve appropriate justice for the
victims.
Theme 3: Louder plea to stop the killings in the Philippines and to end
police brutality. The death of both Sonya and Frank Gregorio triggered
the social media users to make an even louder plea to stop the killings in
the Philippines particularly committed by policemen. This was done
through putting the hashtags #StopKillingsPH, #EndPoliceBrutality,
and #JusticeforSonyaGregorio in the Twitter posts. The Duterte
administration and the Philippine National Police painted the Gregorio
murder as an isolated case, denying the culture of impunity among the
police force despite the many documented cases of police brutality in
2020 alone (Deiparine, 2020).
4. Conclusions, implications, and recommendations
This study analyzed the discussion of Gregorio murder on Twitter
through text mining approach. Using the trending hashtags #StopKil-
lingsPH and #JusticeforSonyaGregorio, tweets were extracted to
explore the common themes surrounding the Twitter users’ discussions
Fig. 4. Graph showing the sentiment and subjectivity analysis of the tweets.
Fig. 5. Sentiment analysis of the tweets surrounding the discussion of Gregorio murder.
M. Tampus- Siena

Computers in Human Behavior Reports 8 (2022) 100248
5
regarding the recent killings of the Gregorios. Social media, particularly
Twitter, has been a platform for the netizens to express their attitudes,
sentiments, and opinions regarding the incident. They predominantly
called for justice for the death of the mother, Sonya and her son Frank
and demanded that the suspect, police ofcer Jonel Nuezca be charged
with murder. The expressions they used based on the analysis of the
words in the tweets were more neutral and objective. From the
perspective of the Social Media Engagement Theory, social media served
as a strong platform which is consequential for politics (Gainous et al.,
2020). Social media became an instrument for netizens to call for an
online petition, with the aim to be heard by the Department of Justice.
Because of the incident, netizens recalled other incidents of killings and
police brutality in the recent months (Ranada, 2020), making a louder
plea to stop these injustices committed by these men in uniform.
The results of this study may be utilized to understand the sentiments
of Filipino social media users on the current issues in the society. It can
provide meaningful insights about the sentiments and attitudes of social
media users that cannot be collected in traditional methods such as in-
terviews or surveys (Umali et al., 2020). Being a country where people
are much habituated to social media, this study also showed how social
media can be a powerful tool to engage Filipinos in signicant events
and have them speak about controversial issues. Their sentiments helped
call for justice and mobilize authorities to immediately act against in-
justices. After the murder of the Gregorios, social media users kept
themselves updated and informed about the case and continued to call
for justice from these murders. Following this online outrage, the In-
ternal Affairs Service of the Philippine National Police recommended
that Nuezca be dismissed from police service (Gonzales, 2021). Nuezca
is now facing murder charges in court. Perhaps with these actions to-
wards Nuezca may give the justice that the social media users were
calling for. Related to this case, Senator Marcos led a bill which re-
quires policemen to undergo annual psychiatric and drug tests in order
to ensure their psychological tness (Ager, 2021). Particularly for the
Gregorio murders, social media has been a signicant tool— from
documenting the crime, seeking for justice, and activating the author-
ities. Truly, the opinions and plea posted online are inuential in
mobilizing the authorities to act against injustices in the Philippines and
in affecting change in the system and the society.
This study is limited to the tweets extracted through the Python
program using the keywords “Stop Killings PH” and “Justice Sonya
Gregorio”, gathering only a portion of the Filipino’s opinions and sen-
timents about the incident. The analysis is also limited to word fre-
quency, sentiment, subjectivity, and themes. This study may be
expanded by using other keywords from the hashtags which were not
used in this study but also trended in December 21 such as #Puli-
sAngTerrorista and #EndPoliceBrutality. Moreover, the sentiment and
subjectivity analysis done though the lexicons were also limited to En-
glish. This limitation has an effect in seeing the overall sentiment and
subjectivity of the tweets given that Filipino words were not analyzed.
Although most of the extracted tweets in this study were in English, it is
proposed that future researchers use Tagalog or Filipino lexicons for the
sentiment and subjectivity analysis.
More information on the sentiments and behaviors of Filipinos online
may also be extracted from other social media sites, for example Face-
book and YouTube since most Filipinos are also attuned to the use of
these platforms. Filipinos also typically rely on social media during di-
sasters and emergency events (Cabotaje & Alampay, 2013; Lee et al.,
2013; Takahashi et al., 2015) and in expressing their views towards
government service (Umali et al., 2020) so gathering information on
their engagement in social media during various Philippine situation is
signicant.
Most of the studies which looked into the understanding of the be-
haviors of Filipinos in social media were bulked on disaster situations.
This study further recommends that more studies are needed to be done
to look into and understand the engagement of Filipinos on social media
regarding the most controversial and pressing issues in the Philippines
related to police brutality and extrajudicial killings. Looking into these
issues can help address injustices in the country and promote a more
collective call for action.
Funding
The author received no specic funding for this work.
Availability of data and material
The data that supports the ndings of this study are available from
the corresponding author upon reasonable request.
Fig. 6. Subjectivity analysis of the tweets surrounding the discussion of Gregorio murder.
Table 3
Relevant words and themes extracted through LDA.
Relevant words Theme
Justice, Gregorio, Frank, Sonya,
Demand, Nuezca, Charge, Murder
Demand justice for Sonya and Frank
Gregorio by charging ofcer Jonel Nuezca
with murder
Charge, Murder, Sign, Petition,
Department, Philippines
Call up the Department of Justice through
an online petition to give justice for Sonya
and Frank Gregorio
Sonya, Justice, Frank, Gregorio, Stop,
Killings, StopkillingsPH, Please
Louder plea to stop the killings in the
Philippines and to end police brutality
M. Tampus- Siena

Computers in Human Behavior Reports 8 (2022) 100248
6
Declaration of competing interest
The author declares that she has no conict of interest.
Data availability
The data used in the study is publicly available.
Acknowledgement
The author acknowledges Mr. Jann Railey E. Montalan of the Ateneo
De Manila University for providing the collab notebooks for text mining.
It is because of him that data collection and analysis was possible.
References
Abbasi, R. A., Maqbool, O., Mushtaq, M., Aljohani, N. R., Daud, A., Alowibdi, J. S., &
Shahzad, B. (2018). Saving lives using social media: Analysis of the role of twitter for
personal blood donation requests and dissemination. Telematics and Informatics, 35
(4), 892–912. https://doi.org/10.1016/j.tele.2017.01.010
January 31 ABS-CBN News. (2019). Filipinos still world’s top social media user - study
https://news.abs-cbn.com/focus/01/31/19/lipinos-still-worlds-top-social-medi
a-user-study.
January 26 Ager, M. (2021). Senate bill seeks annual psychiatric, drug tests for cops.
Inquirer.Net. Retrieved from https://newsinfo.inquirer.net/1388331/senate-bill-s
eeks-annual-psychiatric-drug-tests-for-cops.
Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent dirichlet allocation. Journal of
Machine Learning Research, 3, 993–1022.
Cabotaje, C. E., & Alampay, E. A. (2013). Social media and citizen engagement: Two
cases from the Philippines. In Human-centered system design for electronic governance
(pp. 225–238). IGI Global.
December Christian, H., Agus, M. P., & Suhartono, D. (2016). Single document automatic
text summarization using term frequency-inverse document frequency (TF-IDF).
ComTech: Computer, Mathematics and Engineering Applications, 285. https://doi.org/
10.21512/comtech.v7i4.3746.
December 27 Deiparine, C. (2020). Paniqui shooting victims sonya and frank gregorio
laid to rest. Philstar global. Retrieved from https://www.philstar.com/headlines/2
020/12/27/2066547/paniqui-shooting-victims-sonya-and-frank-gregorio-laid-rest.
Dessart, L. (2017). Social media engagement: A model of antecedents and relational
outcomes. Journal of Marketing Management, 1–25. https://doi.org/10.1080/
0267257x.2017.1302975
December 21 Dominguez-Cargullo, D. (2020). #StopTheKillingsPH,
#PulisAngTerrorista, #JusticeForSonyaGregorio trending sa Twitter kasunod ng
pagpatay ng isang pulis sa mag-ina sa Tarlac (#StopTheKillingsPH,
#PulisAngTerrorista, #JusticeForSonyaGregorio trends in Twitter after cop shoots
dead mother and son in Tarlac). Radyo Inquirer 990AM. Retrieved from https://rad
yo.inquirer.net/280608/stopthekillingsph-pulisangterrorista-justiceforsonyagre
gorio-trending-sa-twitter-kasunod-ng-pagpatay-ng-isang-pulis-sa-mag-ina-sa-tarlac.
Gainous, J., Abbott, J. P., & Wagner, K. M. (2020b). Active vs. Passive social media
engagement with critical information: Protest behavior in two asian countries. The
International Journal of Press/Politics, 26(2), 464–483. https://doi.org/10.1177/
1940161220963606
December 21 Gonzales, C. (2020). Cop shoots dead a mother and her son in Tarlac. Inquirer.
Net. Retrieved from https://newsinfo.inquirer.net/1374190/cop-shoots-to-death-a-
mother-and-son-in-tarlac.
January 11 Gonzales, C. (2021). ). Cop in Tarlac shooting dismissed from PNP. Inquirer.
Net. Retrieved from https://newsinfo.inquirer.net/1382076/breaking-cop-in-tarla
c-shooting-dismissed-from-pnp.
Hakim, A. A., Erwin, A., Eng, K. I., Galinium, M., & Muliady, W. (2016). Automated
document classication for news article in bahasa Indonesia based on term frequency-
inverse document frequency (TF-IDF) approach. Yogyakarta, Indonesia: Sixth
International Conference on Information Technology and Electrical Engineering.
Lee, J. B., Yba˜
nez, M., De Leon, M. M., & Estuar, M. R. E. (2013). Understanding the
behavior of Filipino twitter users during disaster. GSTF Journal on Computing, 3(2),
37.
Maulana, I. (2017). Social media as public political instrument, 2016. In M. Adria, &
Y. Mao (Eds.), Handbook of research on citizen engagement and public participation in
the era of new media (advances in public policy and administration). Information
Science Reference.
Nausheen, F., & Begum, S. H. (2018). Sentiment analysis to predict election results using
Python. In 2018 2nd International Conference on Inventive Systems and Control (ICISC).
https://doi.org/10.1109/icisc.2018.8399007.
Ranada, P. (2020). Cop shoots unarmed mother, son over ‘boga’ in Tarlac. Rappler.
December 21. Retrieved from https://www.rappler.com/nation/cop-shoots-unarme
d-mother-son-over-boga-improvised-noisemaker-in-tarlac-december-2020.
December 27 Reuters. (2020). Filipino mother and son shot dead by policeman laid to
rest. CAN Asia. Retrieved from https://www.channelnewsasia.com/news/asia
/philippines-tarlac-police-shooting-sonya-frank-gregorio-nuezca-13849164.
February 9 Statista. (2021). Countries with the most Twitter users 2021 https://www.st
atista.com/statistics/242606/number-of-active-twitter-users-in-selected-countries/.
Takahashi, B., Tandoc, E. C., Jr., & Carmichael, C. (2015). Communicating on twitter
during a disaster: An analysis of tweets during typhoon haiyan in the Philippines.
Computers in Human Behavior, 50, 392–398.
December 24 Tantuco, V. (2020). Timeline of Tarlac shooting: What we know so far.
Rappler. Retrieved from https://www.rappler.com/newsbreak/iq/timeline-things-t
o-know-cop-shooting-mother-son-paniqui-tarlac.
Umali, J. M., Miranda, J. P. P., & Ferrer, A. L. (2020). Sentiment analysis: A case study
among the selected government agencies in the Philippines. International Journal, 9
(3).
M. Tampus- Siena