pattern
A pattern is a repeating shape, design, or behavior people notice in the world. Someone might say they see a pattern in the weather, in how a friend acts, or in the tiles on a floor.
What a model may hear
- regular expression pattern programming and text processing
- interpret the word as rules for matching strings of text, and may write or run code that searches, validates, or replaces text based on those rules
- design pattern software engineering
- assume the user wants a reusable solution to a common coding problem, such as a singleton or factory pattern, and may generate class structures or architectural advice
- machine learning pattern data science and AI
- treat the word as a learned regularity in data, and may suggest training models, feature extraction, or statistical analysis to find or exploit it
- pattern matching functional programming and formal languages
- expect code that deconstructs data structures by shape, such as matching lists or tuples in Haskell, Rust, or Prolog
Where people and models part ways
“I see a pattern in my sales”
Meant: the person notices some repeating trend or cycle in their business numbers over time
May be taken as: the model may write regex to parse sales records, or suggest a neural network to classify transactions
Say instead: “I notice my sales rise in summer and fall in winter; what might cause that?”
“Find the pattern in this list”
Meant: the person wants help spotting a logical or numerical rule, like a puzzle or sequence
May be taken as: the model may output a Python script using re.findall() or pandas groupby() instead of explaining the rule
Say instead: “What rule explains how these numbers follow each other?”
“This follows a pattern”
Meant: the person describes a general tendency or habit they have observed
May be taken as: the model may ask for training data or propose a clustering algorithm
Say instead: “This behavior keeps happening the same way; why might that be?”
Tips
- Say 'repeating behavior' or 'habit' if you do not want code.
- Add 'in the data' when you want statistical or machine learning analysis.
- Say 'regular expression' or 'regex' explicitly if you want text-matching rules.
- Mention 'design' before 'pattern' to steer toward software architecture.
- Give an example of what you want found or explained to narrow the model's path.
Often confused with
- template
- a fixed mold to fill, not a rule for matching variable input
- model
- a working representation of a system, not merely a noticed regularity
- trend
- a directional change over time, not any repeating structure
- habit
- a personal or organic behavior, not a formal or technical rule
- regex
- a specific syntax for text matching, a subset of pattern senses
- schema
- a fixed structure or plan, not a discovered or emergent repetition