Using AI in preventing child grooming

The internet has undoubtedly made the world a smaller place with more opportunities for people to connect. It’s also created new risks and dangers. One such threat may come from a more experienced adult hunting down the most vulnerable members of society – children, grooming them into sexual activity through chats or social media.
Many people are unaware of the dangers of child grooming. What is it? Well, to put it simply, it’s when an adult poses as a minor online in order to get someone else’s child to meet up with them for sex. It can lead to serious consequences. The good news is that there are measures we can take that may help prevent this type of crime from happening.

Solution?

For example, Facebook recently partnered with the UK police force on an AI program called Photo DNA which identifies images of child exploitation before they’re shared or uploaded.

Technology is a great resource for parents who find themselves in abusive situations. Technology can be used to determine if your child’s been abused, and then report it accordingly!

Developing an algorithm that can detect child grooming and other malicious intent before they have the chance to take place. 

The artificial intelligence that prevents child grooming

There are a few different options when it comes to dealing with this problem with their own pros and cons. The first option is probably the easiest but also has its drawbacks in terms of effectiveness and cost, whereas choosing between these two other solutions will come down largely on what you prefer: long-term results or quick fix right now. 

The easiest way to set up an application on AWS Comprehend is by connecting it with their services. They offer models that work well for the task at hand, but if they aren’t doing their job then you can always create new ones and train them using machine learning techniques before deploying them onto this platform as needed! Additionally, there are often various types of NLP (neuro-linguistic processing) available to deploy on the service named above.

2 top AI solutions comparison

Price calculation:


Example 1 – Analyzing Comments

 

In my example scenario, the application uses Amazon Comprehend to analyze comments on the website. The app has received over a year 10,000 comments that are 550 characters each. 

Total charge calculation:

Size of each request = 550 characters

Number of units per request = 6

Total Units: 10,000 (requests) x 6 (units per request) = 60,000

Price per unit = $0.0001

Total cost = [No. of units] x [Cost per unit] = 60,000 x $0.0001 = $6.00

 

Custom Entities API

Example 2 – Analyzing Customer Comments using the Custom Entities API

Assume you want to train a custom entity model to automatically extract custom terms from customer feedback that comes in from your website. The training operation takes 1.5 hours for analyzing 10,000 pieces of customer feedback that are 550 characters each. Your plans for keeping this model are approximately for a month. Pretend you use this service for over a year and you’re not able to have a free tier. The output should have been witty

Total charge calculation:

Size of each request = 5,500,000 characters

Number of units per request = 55,000 units [5,500,000 characters ÷ 100 character per unit]

Price per unit = $0.0005

Total cost for units = $27.5 [55,000 units x $0.0005]

Total hours for model training = 1.5 hours

Price per hour = $3

Total cost for model training = $4.5 [1.5 hours x $3]




Number of months for model management = 1 month

Price per month = $0.50 

Total cost for model management = $0.50 [1 month x $0.50]

Total cost = $37 [$27.5 + $4.5 + $0.50]

While the AWS SaaS model is useful for the deployment and management of a text analytics infrastructure, it could be that your particular case requires more flexibility. Nothing could be compared to a custom-made Neural Network. You get it all in one place and can seamlessly combine the features of other networks with ease! The advantages are that you can have few NN models or have a Complicated model structure with custom algorithms and still achieve greater accuracy. 

The disadvantage of this solution is increasing level of complexity. The problem is that you have to keep all infrastructure related. This can be challenging but it’s worth it.  

The price for a GPU server with heavy lifting performance varies from as low as 150$/month to 800$/month depending on the performance needed.

After all, every decision needs to be carefully analyzed.

 

 

We have a wide range of AI options for every need and budget. If you’re unsure what will work best, contact our customer support team info@datapipesoft.com.

 

 

References:
https://aclanthology.org/W13-1607.pdf

https://ntnuopen.ntnu.no/ntnu-xmlui/bitstream/handle/11250/2623143/no.ntnu:inspera:2526935.pdf?sequence=1

https://medium.com/omdena/a-simplified-nlp-model-to-prevent-child-and-sexual-abuse-in-organizations-ccc81f4668c

https://link.springer.com/chapter/10.1007/978-3-319-13734-6_30

 

 

 

 

Leave a Reply

Your email address will not be published. Required fields are marked *