# AI farming game FAQ

<figure><img src="https://229443051-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fbd1NG2G8KKGj95naozcg%2Fuploads%2FY0woRBq3zpc7baE1w1KE%2FTutonow%20(1).png?alt=media&#x26;token=cdd3b1e2-ed61-43ba-82e6-cfeaf147fd20" alt=""><figcaption></figcaption></figure>

<mark style="background-color:green;">The objective of our game is to correct text on pictures of cannabis brands and strains.</mark> We randomly select an image from our curated content database and use OCR technology to identify all the words in the picture.&#x20;

These words are presented to game players for validation. To validate an image, each word must be verified once, and then undergo an additional three validations to confirm the image. Once a word is validated, it will not reappear, even if the image is still unconfirmed.&#x20;

When the same image is presented again, previously processed words are retrieved from the database. If a user corrects a word, the corrected version is visible to other players, and if accurate, it will be confirmed, increasing the word confirmation count. If a word is deemed incorrect and adjusted, the confirmation count resets to zero.&#x20;

However, if a word is skipped, it does not affect the existing number of confirmations for that word. The system only validates an image if multiple users confirm that the initial labeling process was correct. Once an image is validated, it can be used to train the CBC AI.&#x20;

<figure><img src="https://229443051-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fbd1NG2G8KKGj95naozcg%2Fuploads%2FtfFBWY1GOeStZmrF7KIj%2FScreenshot%202023-03-29%20at%2012.02.31%20PM.png?alt=media&#x26;token=ec20a582-ce22-4ff1-9910-58a4f43903f9" alt=""><figcaption></figcaption></figure>

<mark style="background-color:green;">**To find the cannabis brand on a product label, look for the following information:**</mark>

1. Logo: A brand's logo is often a prominent feature on the label. It is a visual representation of the brand and helps customers identify the product.
2. Brand name: The brand name is typically displayed prominently on the label. It might be written in a distinctive font or color to make it stand out.
3. Contact information: Some labels will also include the brand's website or social media handles, which can help you identify the brand and learn more about their products.
4. Product description: The product description often contains the brand name as well, especially if the brand offers various product lines or strains.

<figure><img src="https://229443051-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fbd1NG2G8KKGj95naozcg%2Fuploads%2F4N7J2t7ICL21FgfvSNGf%2FBrands.png?alt=media&#x26;token=1f5ee7a7-c9f7-45d6-94fb-4587fc474aea" alt=""><figcaption></figcaption></figure>

<mark style="background-color:green;">**To find the strain name on a cannabis label, look for the following:**</mark>

1. The product name or title: Many cannabis products will have the strain name incorporated into the product name or title, such as "Blue Dream Pre-Rolls" or "Gorilla Glue #4 Flower."
2. Strain information section: Some labels may have a dedicated section or field for strain information, where the strain name is clearly listed.
3. Description or details: If the strain name isn't prominently displayed, it might be mentioned in the product description or additional details on the label.

<figure><img src="https://229443051-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fbd1NG2G8KKGj95naozcg%2Fuploads%2FL02FWyy7smkGlpXDvq8n%2FStrains.png?alt=media&#x26;token=c03a6e78-6ecc-4330-a168-15c6ac5821ea" alt=""><figcaption></figcaption></figure>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://cbcai.gitbook.io/cbc-ai/ai-farming-game-faq.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
