AI/ML for identification of next attack on Hindu temples

Motivation:

Srila Prabhupada asked us to peacefully chant the holy names of Krishna and Rama and ultimately achieve liberation and go back to Godhead. The envious people around us have a different plan; they want to finish us before we have an opportunity to purify our soul. Hence we need to rise up to the Ksatriya standard and develop strong defense mechanisms to chant holy names of Krishna and Rama.

ML example 1: Student violence prediction

Hare Krishna, Hare Krishna, Krishna Krishna, Hare Hare

Hare Rama, Hare Rama, Rama Rama, Hare Hare

Data Example 1: An escalation in Israel:

This shows how a big event is triggered by small events and can be predicted via a advanced warning system.

There is no one cause for this escalation. Rather it results from a collection of forces and strategic interests converging. Like the epic art of Middle Eastern story-telling, the singular “umbrella” theme of escalation is actually the product of many separate sub-tales woven into other tales which align into a shell or framework story. In this case, that unifying shell tying these separate tales together represents a very real moment of danger.

The signs of escalation were building for weeks. In early April, there was a sudden escalation of attacks on Israeli Jews, many of which were serious and violent enough to result in hospitalization. As the Palestinian Media Watch, and FLAME – an organization dedicated to accuracy in media – note, the Palestinian official media organs started broadcasts of highly inflammatory and bloody rhetoric starting on April 2.[1]?Two particularly disturbing attacks, one a beating by three Arab youths of a Rabbi in Jaffa, in the southern part of Tel Aviv, and another when an Arab spilled boiling liquid on a Jew entering the Old City of Jerusalem, were followed by violent Arab demonstrations when police attempted to arrest the perpetrators.[2]

Non-Violent advanced warning systems needed:

We live in the age of predictive analytics based on tools like IBM Watson, the Azure, GCP and AWS algorithms. Many Hindus world-wide are studying in advanced universities like MIT, Stanford, CMU, UMich with their strong Artificial intelligence and Machine learning programs. I urge these students to build a ML model with deep-learning and take the data from the last 40,000 temple attacks from history text books and develop a model to predict the next attack on a Hindu temple. Atleast the last 100 temple attack data can form the basis for this algorithms - will make a good PhD thesis and your will be remembered for a long time. As chaos theory suggests there is some pattern even in Chaos and can be understood and leveraged.

Features in a Hindu temple Attack Prediction (HTAP) system:

1. Economic factors: Sudden land price increases in the Hindu neighborhood prompting envious people to attack
2. Festivals: Joyous festivals will make envious people more envious
3. Wealth increase: Hindus around the world are usualy hardworking and wealthy - need I mention Satya Nadella, Sundar Picchai, Arvind Krishna, etc. Pockets of wealth increase are documented by local country financial databases
4. Advertising: keywords in advertisements such as "diwali crackers" "Fabindia" "Dabur Lesbian Ads" "Shariah" "Kalma" etc in social media
5. Laxity amongst Hindus: With too much comfort comes laxity, people slide from Kshatriyas to Sudras. Laxity indices can be measured by second and third generation Hindus not pursuing complex jobs but easy route such as Pizza delivery, Subway counter clerks, etc.
6. Maturity of Temple security protocols and equipment
7. Key events - like the U.S. withdrawal from Afghanistan, additional troops in Yemen, instability on India Pak Border
8. Recent religious pronouncements
9. Poverty of the area. If the delta between the affluent and the poor exceeds 5X, the results are disastrous. There is a constant threat to the well-being of the affluent especially in a religious charged town
10. Gang activity in the area - local gang's revenue models now include foreign funding and parachuted commandos with advanced training.
11. Social media signals. Before any major activity there is a flurry of keywords in the social media. APIs to connect to social media at Petabyte scale data analysis can be expensive
12. New push for religious conversion - whether it is the likes of Dawood Ibrahim or some new evangelist, monitoring their movements becomes important for the model

...a big data and ML technique for behavior analysis and crime prediction is presented. This paper discusses the tracking of information using big data, different data collection approaches, and the last phase of crime prediction using ML techniques based on data collection and analysis. A predictive analysis was conducted through ML using RapidMiner by processing historical crime patterns. The research was mainly conducted in four phases: data collection, data preparation, data analysis, and data visualization. It was concluded that big data is a suitable framework for analyzing crime data because it can provide a high throughput and fault tolerance, analyze extremely large datasets, and generate reliable results, whereas the ML based na?ve Bayes algorithm can achieve better predictions using the available datasets. [1]

More can be written on this topic and this is a just a beginning of a great defense mechanism against Hindu/ISKCON temples. The attacks will not slow down, throughout history this has happened and there is no reason to think this will slow down.

Opinions expressed are personal and not the organization the author works for.

References:

1. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8081790/
2. https://centerforsecuritypolicy.org/anatomy-of-an-intentional-escalation-israels-approaching-hot-summer/
3. https://www.nap.edu/read/4422/chapter/5#218