Audit Course

Social Media And Analytics

Audit Course Report submitted to Savitribai Phule Pune University, Pune

In partial fulfillment for the awards of Degree of Engineering in

Computer Engineering
Submitted by

Tejas Hiralal Lodaya – 72149915F

Under the Guidance of

Prof. S. B. Ambhore

Designation of Guide:
Assistant Professor

Sep 2023-24

Department of Computer Engineering

SNJB’s Late Sau. Kantabai Bhavarlalji Jain, College of

Engineering, Chandwad, Dist. Nashik

 CERTIFICATE

Department of Computer Engineering

SNJB’s Late Sau. Kantabai Bhavarlalji Jain, College of Engineering, Chandwad,
Dist.
Nashik

This is to certify that, Tejas Hiralal Lodaya 72149915F have successfully

completed the Audit Course entitled “Social Media And Analytics”
under my guidance in partial fulfillment of the requirements for the finalYear of
Engineering in Computer Engineering under the Savitribai Phule Pune University
during the academic year 2023-2024.

Date : ……………….
Place: Chandwad

Sign of Guide HOD Principal

Prof. S. B. Ambhore Dr. K. M. Sanghavi Dr. R. G.Tated
 Acknowledgements
With a deep sense of gratitude, we would like to thank all the people who have lit our path with
their kind guidance. We are very grateful to these intellectuals who did their best to help during
our project work.

It is our proud privilege to express a deep sense of gratitude to Prof R.G.Tated, Principal of
SNJB’s LS KBJ COE, Chandwad, for his comments and kind permission to complete this project.
We remain indebted to Dr. K. M. Sanghavi, H.O.D. Computer Engineering Department for his
timely suggestion andvaluable guidance.

The special gratitude goes to Prof. S. B. Ambhore for excellent and precious guidance in
completion of this work. We thank all the colleagues for their appreciable help for our working
project. With various industry owners or lab technicians to help, it has been our endeavour
throughout our work to cover the entire project work.

We are also thankful to our parents who provided their wishful support for our project
completion successfully. And lastly, we thank our all friends and the people who are directly or
indirectly related to our project work.

Tejas Hiralal Lodaya

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

Introduction
This chapter looks at how to use Social Media Analytics to inform campaigns, strategies and
engagement assessments. It helps you use insights from Social Media data to get a better
understanding of how to engage communities and adjust your programmes.

In this Guide, the term “Social Media Analytics” refers to the systematic computational analysis of data or
statistics from a Social Media account/channel. Typical SM analytics include for example number of individual
engagements (retweets, likes, comments) and audience demographics (by age, gender, location).

Social Media is by definition a real-time means of communication. Regardless of whether you post something
today and I read it tomorrow, people using Social Media expect an almost instant response.114 For this reason, it is
important that the admins of Social Media are able to respond immediately and adapt their channels in near real-
time. Analytics are one of the best ways to keep track of fast-moving, real-time conversations, providing data and
analysis that can inform engagement strategies, community management and management of reputational risk.

Social Media Analytics for Protection can be used for various purposes:

1.Situation Analysis

Often Social Media platforms have embedded tools that perform some basic analysis of the data on your channel.
Depending on the platform, additional analytical capacity can be built into the channel or it may be possible to
extract data with an Application Program Interface (API). You should always have a way to track and measure your
Social Media presence, no matter what you are using it for.

Data held by Social Media companies can help UNHCR and partners better understand key issues related to
a situation, such as: the demographics of certain groups, their geographic distribution, and displacement
trends; public sentiment; and who UNHCR is reaching on Social Media (which helps us assess if we are
reaching the right/targeted audience). Such data contributes to improved situational awareness and response,
and we can also use it to conduct our Social Media Situation Analysis (see Chapter 1). Remember that situation
analysis does not involve or require collecting personal data.

114 According to a study by The Social Habit, 42% of consumers expect a response on Social Media within 60 minutes. On
Facebook, 85% think up to six hours is reasonable while on Twitter, 64% want an answer in an hour. Source here

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

Example

This paper, published in 2017, is the result of an experimental project conducted by the UN Global Pulse and
UNHCR’s Innovation Service. It used data from Twitter to monitor protection issues and the safe access to
asylum for refugees and migrants in Europe. The project looked at interactions among refugees, between refugees
and host communities, and between refugees and service providers along the way to Europe. Find more
information here.

2. Knowledge creation and Transfer

A wide range of datasets generated by Social Media analytics can be combined and analyzed to generate new
insights that can be used by decision makers. This is particularly true when Social Media data is combined with
other sources of data to generate a more complete picture of the issue at hand.

Example
3. Service Design and Delivery
In 2014, the United Nations Millennium Campaign and the Water Supply and Sanitation Collaborative Council
Social Media
partnered analytics
to deliver can open up previously
a comprehensive advocacyinaccessible data sets, drive
and communication improving decisionsTheir
on sanitation. aboutefforts
response strategies
gelled with and
helping
the UNtoDeputy
guide service delivery
Secretary based
General’s onto
Call reliable
Actionevidence.
on Sanitation to increase the number of people with access to
good sanitation. Global Pulse analyzed Social Media to provide an insight on the baseline of public engagement
and explore ways of monitoring a new sanitation campaign. By showing how the volume and content of public
discourse
Example about sanitation changed over time, the study made it possible to monitor the reach and effectiveness of
the communications campaign. Find more about this project here.
Between May and August 2016, when Zika was a big issue in Brazil, Facebook and the data consultancy firm
ActionSprout conducted surveys of local Facebook users and analyzed what they were saying to better
understand the public conversation about the virus. The two organizations anaysed the data to find out who was
posting information about Zika and what they were sharing to determine their concerns. This allowed UNICEF
Brazil to tailor their Social Media ads and provide better services to communities to help them with Zika
prevention.

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

4. Prediction and Forecasting

New predictive models made possible by access to Social Media datasets can help UNHCR conduct forecast
analysis to anticipate problems or avert crises.

Example
5. Impact Assessments and Evaluation
In 2017, University of Warwick researchers looked at photographs and key words posted online to see if they
could signal weather risks developing in specific locations. For example, posts about water levels rising could
Access to authorities
alert the Social Media to adatasets
potential can helpThe
flood. UNHCR monitor
researchers and evaluate
tracked photos the
andimpact
videos of government
with tags (suchrefugee
as river,policies
115
and UNHCR
water responses.
or landscape) on Thus,
Flickrthe design 2004
between of programs
and 2014. andTheservices can showed
findings be constantly improved.
that tracking certain words used on
Social Media around the time of an extreme weather event allows information to be collated to predict which
areas will be affected. You can work out how serious the threat to life and infrastructure will be and create an
While
early some
warningof the examples
system Data116 projects, the analysis of Social Media data does not necessarily
above are Bigaccuracy.
of unprecedented
Example
require Big Data. Inevitably, when digital platforms are used, data is generated both by UNHCR and end-users
and shared among various parties, including the platform provider. The data can be directly generated by the
In 2014, to coincide with the United Nations Climate Summit, UN Global Pulse developed a real-time Social Media
interactions on Social Media and/or come in the form of metadata i.e. data about the data generated.
monitoring system to explore online discourse about climate change. The monitor was accessible to the public
and daily analyzed tweets in English, Spanish and French to show the volume and content
of comments about climate change in relation to the economy, energy, etc. By comparing interest levels between
115 See Chapter 11 for more information on how to use Social Media for M&E of UNHCR Social Media activities.
topics
116 The and regions,
UN defines Big and
Data monitoring the success
as the large amount of climate-related
of passively communications
collected data deriving and events,
from everyday interactions with the monitor could
digital
measure
productsawareness, further public
or services, including engagement
mobile phones, andand
credit cards support climate policy-making. Access the Twitter monitor
Social Media.
here.

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

With data analytics, certain data protection concerns need to be considered:

• Big Data and data analytics may not involve the processing of personal data or personally identifying data.
But with enough information, it may be possible to generate a profile of a user type and his/her habits;
• On Social Media platforms, consent, when sought, is often not fully informed or specific, and data
submitted for one purpose may be reused out of context for unrelated purposes. So we have to be very
careful about what Consent means and make sure it covers the range of the data we are managing and
processing.

While using information from Social Media can support our protection objectives, it is important to know the risks
of doing so. People can be harmed as well as helped by data (and decisions based on data), so we must ensure the
necessary measures are taken to prevent and/or mitigate these risks. In the case of personal data, this can be done,
for example, by conducting a Data Protection Impact Assessment (DPIA)117 (see Chapter 2).

© UNHCR/Gordon Welters

It is also important to remember that data from Social Media platforms and Messaging Apps are typically not
representative of a population beyond the users of the particular channel. Lack of coverage, under- coverage,
self-selection and differing response rates all introduce biases when we estimate the extent of interest from
Social Media data. Such biases should be taken into account before undertaking data analysis. Check sample
distributions in demographic variables against population distributions from censuses, registration data and other
representative data sets. Statistical techniques such as multi-level regression with post-stratification (MRP), can be
used to correct differences between the Social Media sample and population distributions.

117 See the template in the Guide for the UNHCR Data Protection Policy and Chapter 5 of the ICRC Handbook on Data Protection in
Humanitarian Action (2nd edition/2020).

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

1.Social Media Analytics:

Key Tips
TIP 1: Become familiar with the insights options of your chosen Social Media platform.

Almost all of them have one. If yours doesn’t, there may be an option to download and analyze your data independently. Make sure you

TIP 2: Understand the operational context and protection challenges and be clear why you are doing the analysis.

The context and purpose will help you decide what types of data — and at what unit of measurement, granularity and frequency — will he

© UNHCR/Genesis Andreira Lemus Guacaroin

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

TIP 3: Use the metrics below to determine ways of improving your intervention, be it content, response time, etc

What is your target audience looking for? How do they engage with and respond to your online activities? Metrics include:
Age, gender and diversity-disaggregated data about your users;
Their overall level of engagement on your channel;
Topics people respond to/topics that attract less attention;
Most popular format of your posts;
Time of the day when your audience is most online.

At this stage, identify performance indicators for the objectives of your Social Media presence (e.g. number of likes and shares) and con

Resources
See Chapter 11 for more on Social Media and Monitoring and Evaluation practices.

TIP 4: Identify the mechanisms, systems, processes and/or decision points into which Social Media data can be int

You can only do this if you have mapped what information is available and being used to design programs/ activities, draw conclusions an

Real-Time Data for Adaptive Management

Real-time data systems employ digital technologies (computers, tablets, mobile phones, sensors, etc.) and specialized software to accele
Source:

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

TIP 5: When trying to meet the needs of different stakeholders, be aware of trade-offs between speed and data qual

There are clear differences between the characteristics of Social Media data and the requirements of quality data. To be of value, the raw

TIP 6: Be agile if you want to meet the needs of adaptive programming.

Because Social Media platforms are built on regular, fast cycles of action, analysis and adaptation, any system they rely on that will also n

TIP 7: Review, adjust, test, iterate.

Study your metrics consistently and regularly. Some of the analytics systems embedded in your Social Media platform will let you crea
independently. No matter how you do it, make sure you do a periodical review of the data, with frequency determined by the purpose of th

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

2. Tools for Data Analytics

Important
When deciding about technology and what tools to use, reach out to colleagues in UNHCR’s:
Division of External Relations, Digital Engagement Service
DIMA Units in Regional Bureaux
UNHCR’s Innovation Service

The table below shows Data Analytics tools embedded into the most commonly used Social Media platforms:

Social Media Channel Data Analytics Dashboard/Tool

Facebook

Facebook Page Insights give you detailed analytics for your Facebook page,
Facebook Page Insights so you can learn how people interact with your content, note what works and
improve your results.

Facebook Audience Insights help you understand your Facebook audience,

Facebook Audience
so you can better target ads and create more relevant content.
Insights

Twitter

Account Home is the Twitter report card, with high-level statistics

tracked month to month. It’s also a gallery of the account’s greatest hits.
Account Home
It has a spotlight of the top-performing tweets and reveals the
influencers in your network.

The Tweet Activity Dashboard is where you can find metrics for every
Tweet Activity Dashboard single tweet. The dashboard tracks how many times Twitter users have seen,
liked, retweeted and replied to each tweet.

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

Instagram

Instagram Insights tell you about followers and the people interacting with an
account on Instagram. They can provide data about the audience, like gender,
age range and location. The system also allows you to see which posts and
Instagram Insights stories the audience engages with most. Insights and metrics about an Instagram
account include paid activity as well. Reach, accounts reached, impressions and
impressions day by day reflect both paid and free activity.

You Tube

The Watch Time Report compiles data from your YouTube channel, the
The Watch Time Report individual videos on the channel and any engagement from YouTube’s mobile
apps.

The Revenue Report shows how much money an account is making using ads
Revenue Report
within a given period, and where the revenue is coming from.

The most important Interaction Metrics are subscribers, likes/dislikes,

comments and shares; in other words, all the ways viewers can tell you
Interaction metrics directly how they feel about your videos. Studying this data will show you
the best way to reach and cultivate relationships with your subscribers.

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

You can also use a third-party tool to analyze your social media data and get insights on your Social Media
presence.118 The table below shows some third-party Data Analytics tools commonly used to analyze SM platforms:

Platform Description

Brandwatch Consumer Research delivers instant insights about an account,services, users and
trends by providing live and historical data from across the web.
Brandwatch is a paid system that:
Brandwatch
• Collects and analyzes billions of conversations;
• Measures the impact of your marketing, content and social activity;
• Uncovers insights to improve your strategy.

Meltwater helps you sift through thousands of news sources using sophisticated search
monitors. A paid system, it finds everything relevant to a customer’s operation, both locally
and globally.

Meltwater
Its media monitoring function offers the world’s largest source base, including 300,000 news
sources from 205 countries in 87 languages. Meltwater also offers dashboards that can be
customized to analyze metrics by date, geography, language and opinion/sentiment.

Pulsar aims to help clients find the story in the data. Their social listening and audience
intelligence platform bring all digital audiences together in one place.
Pulsar Their analytics suite also allows you to gather data-driven Social Media insights by leveraging
the power of AI & data visualization, using their media analytics, trends and social listening
tools.

Sysomos is a paid system. It allows for:

• Social Listening. The social research and monitoring engine has unlimited data and no
cap on the number of queries on over 200 billion social conversations;
• Social Monitoring. You can engage, publish and monitor news and Social Media from
Sysomos one platform;
• Social Influencing. Search 900 million profiles and 60k categories to find and
manage influencers across the world;
• Reports & Services. You can optimize social strategy with access to a wealth of
social reports or build what you need with customized reports and social API
access.

118 The UNHCR Global Social Media Team and the UNHCR GCS Analytics Team have paid contracts with some of these companies and can
advise country offices wanting to invest in them.

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

Third-party tools allow you to achieve a higher degree of customization and work on larger datasets. But if you
want to use them, there are some considerations:

• Language: Not all third-party tools can detect and translate scripts. This may be a problem if the
community uses Arabic, Chinese, Russian or any other language with a non-Latin script;
• Data Protection and Security: When assessing the risks associated with these tools, always engage with
ICT colleagues on data security and with Data Protection colleagues on data protection;119
• Resources: Third-party tools are normally more sophisticated and complex than embedded tools. This
means staff will have to be trained and it could take weeks before you are able to take advantage of all
the functions of the dashboard;
• Customization: Being able to customize third-party tools is what makes them appealing. But users
sometimes choose tools that are too complicated for their needs. Your analysis should be guided by the
needs of the intended decision-makers to ensure it is not more complex or in-depth than what is necessary to
meet those needs.

There is a third way to gather and analyze data and that is via an Application Programming Interface (API). Social
media services like Twitter, Facebook and Google have all opened up their products to outside developers via APIs.
These enable outside parties to build a product or app off an existing service.
TweetDeck for Twitter, MySpace apps, Facebook Connect and WhatsApp’s Turn.io are just some examples of social
APIs. With an API, you can use an existing app, or build one, to take the data stream from a Social Media channel and
reorganize or interact with it on a different dashboard. UNHCR colleagues can reach out to the Digital Engagement
Service in the Division of External Relations for more support.

Example

In “Using Social Media for Research, Monitoring and Evaluation in the MENA Region: World Food Program Case
119 Remember that a DPIA is required when the collection and processing or transfer of personal data is likely to be large, repeated or
Study” DFID
structural (i.e. looked at is
where data how news
shared withabout WFP’s reduction/cancellation
an Implementing Partner or third party overofa food
certaindeliveries to Syrian
period of time). refugees
See more on the spread
across
UNHCR Twitter
Policyin on2014/15. Researchers
Data Protection . from the University of Cardiff loaded 24,000 tweets into Cosmos, a free-
for-research data analysis software developed by the university. They then ran network and frequency analyses over
the sample and tried to classify topics, sentiment and locations or origin.

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

3. Tools for Resource

Maximization
Managing Social Media is time consuming because of the sheer volume of conversations/interactions and the need
to respond quickly. For an organization like UNHCR and large operations, staff may not be able to deal
meaningfully with the flood of requests.

To manage FAQs or handle a large volume of questions, you may think about using a third-party tool that automates
some actions staff would otherwise perform. Below we list some options. Consult regional and HQ colleagues
before deciding to pursue any of them:

Tool Description

Artificial Intelligence is the theory and development of computer systems able to perform tasks
AI normally requiring human intelligence, such as visual perception, speech recognition, decision-
making and translation between languages.

A chat bot is an artificial intelligence (AI) program that simulates interactive human
conversation by using pre-calculated user phrases and auditory or text-based signals. Chat bots
are used frequently on social networking hubs and for instant messaging (IM). They are also
Chat Bots
often included in operating systems as intelligent virtual assistants. A chat bot is also known as
an Artificial Conversational Entity (ACE), a chat robot, a talk bot and a chatter bot or chatterbox.

Interactive Voice Response (IVR) is a technology that allows a computer to interact with
humans through the use of voice and Dual Tone Multi Frequencing (DTMF tones) put in via a
keypad. In telecommunications, IVR allows callers to interact with an organization’s host
IVR
system via a telephone keypad or by speech recognition, after which services can be inquired
about through IVR dialogue. IVR systems can respond with pre-recorded or dynamically
generated audio to direct users on how to proceed.

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

If you are thinking about using any of these tools, bear in mind the following:

• AI or use of algorithms may have serious data protection and privacy implications. A DPIA120 is
required before implementing the system;

• Automated response can be very frustrating for the person at the receiving end, especially if they
are in distress or not familiar with the language used. The best option is always to pair an automated
system with a human who can respond appropriately;

• Not all situations are right for these tools. It is important you design and test these systems with
Persons of Concern and involve them in subsequent monitoring and evaluation;

• Some AI technologies are in their infancy. While they may be promising, are they
proportional to your needs? And have they undergone a security audit and a Data
Protection Impact Assessment (DPIA)?

© UNHCR/Houssam Hariri

120 New technologies or systems, tools, modules, or data processing platforms, including inter-operable or shared databases, which are
perceived or expected to carry inherent privacy risks should, as a matter of principle, undergo a DPIA. This includes for example the
collection of biometric data, cloud storage, big data analytics, artificial intelligence, drones, automated decision- making systems, two-way
communications using social media, smartphones or bulk SMS. See more on the UNHCR, Policy on the Protection of Per
Concern to UNHCR, 2015.

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA ANALYTICS

Resourcing

While Social Media Analytics may not fall directly under the duties of Information Management Officers (IMOs), it
is always a good idea to involve them at the outset, both in developing the SOPs for your Social Media
strategy/project generally and, more specifically, in defining your data analytics strategy for Social Media. IMOs
will be able to advise on data flows, roles and responsibilities, and relevant data processes, methods and tools. They
can also ensure that any relevant links are made between the Social Media project and the operation’s Information
and data management strategy. Also keep in mind that collecting personal data outside the anonymized analytics
provided by each SM platform may put at risk the protection of PoCs and should therefore be carefully evaluated
with the Data Controller.
Information Management (IM) is a specialist function in UNHCR but other staff also have expertise in generating,
analyzing and using data.121

Consider these factors when resourcing for Data Analysis in your Social Media project:

• Involve M&E experts: Monitoring and Evaluation experts can help you understand how to link
protection performance and outcome indicators to Social Media metrics during the analysis. They will be
pivotal in linking your Social Media data analysis to data collected from offline activities. In other words,
they will help you bridge your online and offline protection strategies;
• Don’t reinvent the wheel: The type of technology or tool you choose for data analysis will determine
the resources you will have to put into the system. Look first at data analytics tools embedded into your
Social Media account. If they don’t do the job, contact IM and M&E colleagues to explore how best to
analyse the data for your purpose;
• Think about your purpose to ensure tailored and effective data analysis and use: Sometimes when
a large amount of data is available, we want to use it all to create ever more sophisticated data analysis
products. Make sure you focus on the data you need to make certain decisions (i.e. the purpose and
objectives of the project identified at the outset) and let this guide your analysis. Present the findings of the
analysis in products that are fit-for-purpose to those objectives and meet the needs of the end-users.

121 For more information on UNHCR’s vision for data, see the UNHCR Data Transformation Strategy 2020-2025

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA

Do’s Don’ts

DO always identify what data you have access to DO NOT use metrics that count interactions.
and understand how it will enhance your Quality matters more than quantity. As one of the
programs, rather than duplicating data collection. metrics, you can use the number of “likes” received
but they should not be the sole indication of the
quality of your engagement.122

DO set up a periodical review of the findings of DO NOT use metrics without placing them in the
your Social Media data analysis that involves all local context. Look at offline events or changes in
staff and partners in the project (Protection, policies that could affect the interactions people have
Program, IM, M&E, and others). on your Social Media channel.

DO NOT think of analysis as a stand-alone activity or

DO institute flexible systems that allow you to adapt of Social Media Analytics as a reporting mechanism.
your activities quickly based on the findings of your Data analysis is just one of a chain of steps to meet
analysis. the information needs related
to your objective. You need a decision-making process
that leads to actions.

122 See more on measuring community engagement in

Chapter 10

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-

 AUDIT COURSE REPORT: SOCIAL MEDIA

Checklist
Have you identified your purpose and the end-users of the analysis of Social Media data, and adjusted
your approach accordingly?

Have you familiarized yourself with the data analytics tools that are embedded in your Social Media
channel?

Have you decided which data you need to bring together for analysis in order to support your objective(s)?

Did you involve all relevant colleagues from the outset, including Protection, Program, and IM, as well
as partners and end-users of the findings?

Do you have a clear process and timeline for analyzing the data and presenting the findings?

Have you made sure that you are only collecting the data you need, and never sharing personal information in
the context of data analysis?

Have you involved other colleagues (i.e., in DER in HQ or DIMA in your regional bureau) that can help you
understand how you can use data analytics from your Social Media projects in a safe and responsible
way?

Case Studies

UN Global Pulse and UNHCR Innovation Service, Social Media and Forced Displacement: Big Data Analytics &
Machine-Learning, White Paper, 2017

UN Global Pulse, Analyzing Social Media Conversations to Understand Public Perceptions of Sanitation, 2014

DFID, Using Social Media for Research, Monitoring and Evaluation in the MENA Region: World Food Program
Case Study, 2016

SNJB’S LATE SAU. K. B. JAIN COLLEGE OF ENGINEERING, 2023-