An AI-Based Approach To Monitor, Expose, And Counter Anti-Hindu Hate

Anti-Hindu sentiment has been on the rise in recent years. It is a sentiment that advances, amplifies, and articulates hatred against Hindus and Hinduism.

This is a toxic global phenomenon influenced by legacies of the past. Hinduism and Hindus have been at the receiving end of such hate since colonial times.

This hate sentiment is now known as Hinduphobia, Hindumisia, or Hindudvesha. While those who use the terms are bound by their own conviction, in principle, in practice, and in good faith, there is no difference in what they refer to.

• ‘Phobia’ stands for fear

These words refer to Hindu hate, and the Hindu stakeholders involved have genuine interest in countering this avalanche of hatred.

The rampant support for, and encouragement of, anti-Hindu sentiment is the hallmark of the humanities and social science departments in several universities around the world, as well as in influential media.

Western academia, in particular, is culpable for incubating and advancing this hatred, and the media are effective in spreading it.

Unfortunately, anti-Hindu sentiment has manifested itself in many social media platforms, aided and abetted by disingenuous academics.

Social Media — Convenient Tool To Promote Anti-Hindu Sentiment

Social media has enabled over a billion people to directly engage with a mass audience using interactive technologies. It has been a boon and a bane for civil society.

On one hand, it has empowered responsible users to spread informative and useful content, shaping ideas and interests around the world. On the other, it has armed a section of users to spread hateful and vicious content.

Many hate players seek to exploit historical animosities to provoke and promote hate ideologies. Social media is increasingly used to spread misinformation and instigate violence targeting various political, religious, and social groups.

Hindus have been at the receiving end of this hate mongering on social media. There has been a spurt in toxic hate since 2014, with the election of Narendra Modi as the Prime Minister of India.

Anti-Hindu sentiment flourishes on platforms such as Twitter, Facebook, and Instagram with the support of several anti-Hindu groups, including those directed and mentored by academics.

This has become a menace.

Today, social media is the primary source of news and views for many users around the world (Pew Research Report, 2021). As a result, an increasing number of people are exposed to vicious content, leading to potentially lasting damage to communities, societies, nations, and the world at large. The dialogic nature of social media contributes to the amplification of hate.

Countering Online Hate

Policy makers, media professionals, technologists, and several interest groups have been grappling with the menace of social media hate.

Technology companies have protocols in place to ensure a safer and inclusive environment for their users. But despite the strength of their monitoring effort, the solutions they have are far from satisfactory. They are hardly effective in countering anti-Hindu sentiment.

Rampant misuse of technology is also a cause for concern for many governments in the world seeking to regulate and hold social media companies accountable for their platforms. Some countries are stringent in implementing laws, while some others pay lip-service to combating hate speech online.

Be that as it may, several institutions and researchers have shown keen interest in leveraging technology to confront and counter the growth and spread of online hate. For example, Anti-Defamation League (ADL) has been pursuing an initiative for a few years now, called Online Hate Index (OHI), leveraging deep learning technologies. There are several other independent academic initiatives along the same lines.

They all have faced similar challenges earlier. There is no universal definition of hate, so designing a detection model is a difficult and complex problem. Often, what constitutes hate for one may not be considered hate for another. Hate grammar exists in its own space, context, and time.

Additionally, some of the approaches undertaken were limited by the technologies and techniques available at that point in time.

This was until just a few years ago.

Recent advances in artificial intelligence, deep learning, and natural language processing techniques have provided a fillip to researchers’ efforts. This has galvanised the research community to explore further.

One of the ways automated hate detection may be successful is to avoid building a generic hate detection tool. The grammar of hate is different from society to society and culture to culture. What constitutes hate for Hindus may not be hate for other ethnic groups. So, pursuing a generic solution to detect hate will not be successful.

Instead, a domain-specific approach may offer the best chance of success. With that in mind, an initiative was kickstarted to build an automated tool to detect hatred directed at Hindus on social media, especially on Twitter.

An outcome of such an effort is hindumisia.ai — an initiative that prides itself on taking an analytical approach to countering anti-Hindu sentiment on social media, particularly Twitter.

The tool was launched on 12 March 2022 and is accessible here.

The tool essentially quantifies anti-Hindu sentiment and visualises the same for easier understanding.

Anti-Hindu sentiment is detected by applying a proprietary anti-Hindu hate detection model developed using AI and related techniques. This detection is not generic in nature, but one that is built specifically to detect anti-Hindu sentiment.

Visualisation consists of a Global Anti-Hindu Scorecard® with a real-time Global Anti-Hindu Intensity® monitor and a Global Anti-Hindu Trends® tracker. Users will be able to engage with the tool to view daily and historical trends and share the same on social media for amplification.

Public Vs Private Features

The tool comes with two types of features:

• Public (accessible by all)

• Daily statistics such as the number of anti-Hindu tweets and handles, and real-time 24-hour trend

• Access to specific tweets flagged by the model following a curation process

The initial launch, with basic features, was on 12 March 2022. The tool will evolve in the coming months and years based on use and needs.

How can hindumisia.ai be used?

Hindumisia presents a clear danger to Hindu communities throughout the world. A few Hindu organisations have taken the ownership of countering it by engaging in grassroots advocacy in the United States. A tool such as hindumisia.ai will be a great complement in this regard.

A Tool For Stakeholder Engagement

The tool relies on data or factual information and hence can be used effectively in several ways.

• Inform and educate lawmakers, local law enforcement officials, social media organisations, and educational institutions of the realities of anti-Hindu sentiment through empirical evidence

Actions To Pursue

• Share aggregate and specific information (anti-Hindu tweets) with lawmakers, university administrators, and social media companies (Twitter, in this case) for further action

hindumisia.ai will help amplify Hindu voices, elevate Hindu concerns, draw and sustain attention, promote further research and analysis, and influence action and change.

The tool will evolve in the coming months and years based on the use, needs, and wants of the Dharmic community that is actively combating anti-Hindu hate.

We intend to use the tool to reach out and engage with policy makers, media professionals, law enforcement agencies, relevant government entities, think tanks, and social media influencers for action.

We also hope to provoke and influence the like-minded community of Dharmic business technologists to spawn similar and improved initiatives in the future.

The eventual goal is to promote a safe and secure social media platform for exchange of cultural conversations free of hate and prejudice. hindumisia.ai hopes to play a collaborative and constructive role in this regard.