Introducing NNSwearFilter: The Neural Network Anti-Swear Plugin for Minecraft Servers
Running a Minecraft server is a fantastic way to build a community, but keeping chat clean and family-friendly can be a constant battle. Traditional anti-swear plugins rely on simple word lists, which players easily bypass with creative misspellings, spaces, or symbols. NNSwearFilter changes the game entirely by using a neural network to understand the intent behind a message, not just the exact letters. This plugin is the first of its kind, bringing machine learning directly into your server’s moderation toolkit.
How Does NNSwearFilter Actually Work?
Instead of checking a message against a static dictionary, NNSwearFilter feeds the entire player message into a trained neural network. The network then analyzes the string character by character, looking for patterns that resemble known profanity. It doesn’t just scan for whole words; it examines the structure and similarity of letter sequences. This means that even if a player types something like “F************CK” or “F***_GHE-CK,” the network recognizes the underlying swear pattern and blocks the message. The neural network was trained on a variety of obfuscated and intentionally misspelled swears, making it incredibly difficult to bypass.
The Neural Network API Behind the Scenes
Under the hood, NNSwearFilter leverages a custom neural network API developed by ZombieStriker. This API handles the heavy lifting of pattern recognition, moving through each character and evaluating the likelihood that a substring forms a swear word. The plugin is designed to be lightweight, so it won’t cause noticeable lag on your server. You can explore the full source code and the neural network logic on the GitHub page (ZombieStriker/NeuralNetworkAPI) if you’re curious about the technical details.
What Happens When a Swear Is Detected?
In the current version, when NNSwearFilter identifies a message containing a swear word, it simply prevents that message from being broadcast to other players. The offending player receives a private notification that their message was blocked because it contained inappropriate language. This immediate, silent feedback helps educate players without causing public embarrassment. Future updates may include a censor mode that replaces only the offending part of the message, though this requires the network to know exactly how many characters ahead are part of the swear—a challenging problem the developer is actively working on.
Can Players Bypass NNSwearFilter?
In short, no. The plugin’s neural network is specifically trained to ignore common bypass tactics. Adding extra characters, splitting words with punctuation, or using leetspeak won’t fool the system. The network focuses on the overall pattern of the text, so even heavily obfuscated swears are caught. While no automated system is perfect, NNSwearFilter significantly raises the bar for anyone trying to sneak profanity into chat. If you’re a server admin who wants to add this robust protection without complex configuration, you’ll appreciate how the foxygame.net launcher lets you browse and install server-side mods like this directly from its interface, making plugin management refreshingly simple and modern.
False Positives and Learning Patches
One of the biggest concerns with any automated filter is false positives—innocent messages accidentally blocked. During testing, the developer found that NNSwearFilter produced virtually no false positives in normal server conversation. The only way to trigger a false positive was to deliberately “keymash” random strings of text until the network saw a pattern that resembled a swear. For everyday chat, the accuracy is impressive. However, if a false positive does occur, the plugin supports “learning patches.” Server owners can report the specific phrase, and the developer can update the network to allow that word or phrase in the future. This community-driven improvement loop means the filter gets smarter over time.
Installation and Stats Collection
Installing NNSwearFilter is as straightforward as dropping the plugin JAR into your server’s plugins folder and restarting. It works on Spigot, Paper, and most other Bukkit-based servers. Once running, the plugin uses bStats to collect anonymous usage data, including your server’s system architecture, Minecraft version, and plugin version. No chat data or player information is ever sent. If you prefer to opt out, you can disable stats collection by opening the config file and setting “enableStats” to false. This transparency ensures you remain in full control of your server’s data.
Why Choose a Neural Network Over Traditional Filters?
Traditional word filters are static. Players quickly learn which words are blocked and invent new variations. Maintaining a comprehensive blacklist becomes a never-ending chore. NNSwearFilter adapts to the way people actually type, understanding the intent behind garbled text. This makes it ideal for large public servers, family-friendly communities, or any environment where you want to maintain a respectful atmosphere without micromanaging chat rules. The plugin also frees up your staff from constant moderation, letting them focus on building a better player experience.
Key Advantages at a Glance
- Neural network pattern recognition stops obfuscated swears that word filters miss.
- Zero false positives in normal chat during testing, with a learning patch system for rare edge cases.
- Lightweight and compatible with major server platforms like Spigot and Paper.
- Anonymous stats collection that can be disabled with a single config change.
- Active development with plans for future censor modes.
Final Thoughts
NNSwearFilter represents a smart evolution in Minecraft server moderation. By harnessing the power of neural networks, it offers a level of chat protection that simple blacklists can’t match. Whether you’re running a small private server or a large network, this plugin helps you keep conversations clean without stifling player expression. The developer’s commitment to learning patches and future improvements shows that this is a long-term solution, not a one-off experiment. If you’re tired of playing whack-a-mole with creative swearers, give NNSwearFilter a try—it might just be the last anti-swear plugin you’ll ever need.