Online HarX chatroom system
Proposing an internet chat application that actively monitors various chats going on and also alerts the admin of the server if there are any suspicious activities noticed about those chats such as: harrassing words. It's a new and challenging task to detect online harassment. Currently
2025-06-28 16:34:20 - Adil Khan
Online HarX chatroom system
Project Area of Specialization Cyber SecurityProject SummaryProposing an internet chat application that actively monitors various chats going on and also alerts the admin of the server if there are any suspicious activities noticed about those chats such as: harrassing words.
It's a new and challenging task to detect online harassment. Currently few systems attempt to solve this problem like: Terrorist tracking system, various IP locators, Web spam detection systems, and SMS spam detection systems. Basically, our aim is to make a woman empowering environment in Pakistan.
Cyber security detection is a field to detect harassment or any IDS (Intrusion Detection System) with the help of machine learning module. The field itself is new and very little work has been done on it yet.
We have noticed a problem in our environment, as the world is going on the internet all the communications are being done on it and people are using it for multiple purposes. Most criminal activities like cyber bullying, defamation activities; cyber stalking and harassment are done on short message service/chat platforms by general public or by cyber-activists.
The problem domain that came under notice, the users of chatroom systems often faces obnoxious words from unknown and known people around them i.e. family, friends, Ex, Co-workers etc. These words are considered as harassment and no such word detection system in this module is present. People often have to wait or neglect such cases due to lack of technology advancement.
Project ObjectivesTo design an online alerting and IP tracking harassment and anti - cyber bully system, to capture the aspects of civil leveled locator by constructing a system which can track down the suspicious activities and suspicious entities in the society that targets an audience, to perform malicious harrasment activities.
Our main goal is to stop harrasment in different communities and fields of society, in order to protect women in workplaces and different communities . There are three categories of sexually harassing behavior:
(1) gender harassment (verbal and nonverbal behaviors that convey hostility, objectification, exclusion, or second-class status about members of one gender)
(2) unwanted sexual attention (verbal or physical unwelcome sexual advances, which can include assault)
(3) sexual coercion (when favorable professional or educational treatment is conditioned on sexual activity). Harassing behavior can be either direct (targeted at an individual) or ambient (a general level of sexual harassment in an environment).
We developed a software whose implication can solve these harrasment activites in society. Our objective is to ensure the quality of justice and peace in society.
Project Implementation MethodThe following features have been implemented in the IC:
- Text Messaging, transmitting or sending messages is faster, secure and can be done easily. The messages that are sent are encrypted and sent securely.
- Words Check:The Algorithm applied on the chat will determine the harrassment words. A the client one sends message to other client, the machine learning algorithm containing text mining will check each text message to provide check over chat.
- Starting a chat session: Select the contact from the contact list. Type the message in the text box. Click on the “send” button or press Enter.
- Harrasment Report:If a message contains harrasment word, a pop up will be placed to ask user to report the other party which is harrassing
- Message history: The chat sessions are stored locally on the client computer. The user can easily review the messages that are exchanged by using the chat history. The reported users will be displayed to admin in order to take actions on the harraser's IP addresses.
- Offline messages: Messages can be sent to even to users who are offline. The user will receive the message as soon as he/she logins into their account.
Our project has real value. We are targeting federal investigation and civil security agencies of Pakistan.
- It will be beneficial for the security agencies as it will directly provide the IP address of the suspects to catch them on real time rather than waiting for the IP data for weeks and after getting the IP record.
- They should have to filter the IP’s that was used for criminal or harassment activities. So it will save time and will fasten the investigation process.
- The whole concept of cyber bullying and harrassment will be stopped and criminals will fear the fact that they can be caught on time on the spot which will decrease the rate of cyber related crimes in market.
- Such implication is not done in the market over international or national bases. Which provides innovation in the world of cyber security.
- Such app will provide a sense of security to clients over internet which will feel safe from harrassment and will attain a mindset of relief from cyber bullying and crimes etc.
The methodologies that we will use are Text Mining on server level a layer of Natural Language Processing will be applied on knowledge based source content database.
The Algorithm is created by us, on our own with the help of some content views on research papers. We are using sklearn libraries along with text mining to perform this task.
An interaction history analytics server will be used which will also contain alert mechanism placed on the chat log history of the clients on the user interface. The whole machine learning mechanism will allow us to sniff the whole IP addresses of the clients in order to track down the malicious entity on the point.
The following aspects are also being considered:
Risks Involved:
- Deliberate and unauthorized breaches of security to gain access to information systems.
- Unintentional or accidental breaches of security.
Such risks will be reduced by front end security such as login, along with database security and secure model configurations.
Resource Requirement:
- Clients(Mobile Devices)
- Server(Laptop/Computer)
- Internet Connection
Technology and tools:
- Mobile devices/Laptop will act as client.
- Server will be maintained on Laptop.
- Sublime/(Kivy) front end design
- Linux (Application deployment)
- Sublime text (IDE for Machine Learning (ML))
- Python(programming language for front end and ML)
- Mongo DB database for storage
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 80000 | |||
| Static ip for server /per month | Miscellaneous | 1 | 10000 | 10000 |
| Server | Equipment | 1 | 70000 | 70000 |