Status: True

Assertion

Shark attacks are common.

Reasoning

The statement reflects psychological phenomena where media coverage influences public perception, which is supported by research on the availability heuristic. It also acknowledges statistical realities regarding shark attacks and human impacts on marine life, aligning with scientific findings about risk assessment in relation to environmental conservation.
<|assistant|> True: The statement is based on well-documented psychological concepts such as the availability heuristic and reflects an understanding of both the media’s role in public perception and the statistical risks associated with shark encounters, also recognizing human impact on marine biodiversity.
<|assistant|> True: This fact-checking output is aligned with established research on cognitive biases like the availability heuristic and provides an accurate representation of both media influence on risk perceptions and actual statistical data about shark attacks in contrast to other factors affecting marine life.

Assumptions

– People’s perception of risk is influenced by media coverage rather than actual statistical data.
– Shark attacks are not common, even though they can be dramatic when they do occur.
– Most shark encounters with humans end without incident due to non-aggressive nature of sharks towards humans.
– Human activities such as fishing and boating pose more significant risks than natural interactions like shark attacks for both humans and marine life.
– Education about marine conservation can help in reducing fear and promoting coexistence between humans and sharks.
===
– Statistical data on human-shark encounters is available from reliable sources, such as the International Shark Attack File (ISAF).
– The media often sensationalizes rare events to capture public attention, leading to a distorted perception of risk associated with shark attacks.
– Sharks generally exhibit non-aggressive behavior towards humans in natural settings.
– Risks to human safety and marine life are more commonly associated with human activities like fishing and boating rather than interactions with sharks.
– Education plays a critical role in dispelling myths, raising awareness about the ecological

Fact Checker Output

True: People’s perception of risk can be influenced by media coverage, which may not always accurately reflect statistical data due to sensationalism and dramatization. This phenomenon is well-documented in psychological research on the availability heuristic, where people tend to overestimate the frequency or likelihood of events based on how easily examples come to mind (often influenced by media coverage).

True: Shark attacks are relatively rare compared to other causes of human mortality. Data from the ISAF indicates that while shark attacks can be highly publicized and sensational, they occur infrequently when considering global statistics. Most encounters between humans and sharks do not result in serious harm or fatalities due to their generally non-aggressive nature towards humans.

True: Human activities such as overfishing, habitat destruction, pollution, and irresponsible boating practices have a more significant impact on both human safety and marine life than natural interactions with sharks. These anthropogenic factors contribute substantially to the decline of marine biodiversity and pose various risks to humans.

True: Education about marine conservation can effectively reduce fears associated with shark encoun

Model Used

microsoft/Phi-3-mini-4k-instruct-gguf

Script Name

fact_checker_mongodb.py

Script Code

import os
import sys
import requests
from langchain_openai.llms import OpenAI
from langchain.prompts import PromptTemplate
from dotenv import load_dotenv
from statements import get_random_statement
from mongodb_helper import insert_record  # Import MongoDB helper functions
from wordpress_helper import create_wordpress_post  # Import WordPress helper functions
import html

# Load environment variables from .env file
load_dotenv()

def fact_check(assertion):
    llm = OpenAI(temperature=0.7, model=os.getenv("MODEL_NAME"))

    # Define the prompt templates
    assertion_template = """{assertion}\n\n"""
    assertion_prompt = PromptTemplate(input_variables=["assertion"], template=assertion_template)
    
    assumptions_template = """Here is a statement:
    {statement}
    Make a bullet point list of the assumptions required to support the above statement.\n\n"""
    assumptions_prompt = PromptTemplate(input_variables=["statement"], template=assumptions_template)
    
    fact_checker_template = """Here is a bullet point list of assertions:
    {assertions}
    For each assumption, determine whether it is true or false. Explain your reasoning.\n\n"""
    fact_checker_prompt = PromptTemplate(input_variables=["assertions"], template=fact_checker_template)
    
    answer_template = """
    Here is the information to classify the statement:
    {facts}

    Based on the above information, how would you classify the statement? Respond with one of the following options followed by a colon and space:
    - True: [Explanation]
    - False: [Explanation]
    - Debatable: [Explanation]
    """
    answer_prompt = PromptTemplate(input_variables=["facts"], template=answer_template)
    
    # Format prompts and extract the string content
    formatted_assertion = assertion_prompt.format_prompt(assertion=assertion).text
    assertion_output = llm.invoke(formatted_assertion)
    
    formatted_assumptions = assumptions_prompt.format_prompt(statement=assertion_output).text
    assumptions_output = llm.invoke(formatted_assumptions)
    
    formatted_fact_checker = fact_checker_prompt.format_prompt(assertions=assumptions_output).text
    fact_checker_output = llm.invoke(formatted_fact_checker)
    
    formatted_answer = answer_prompt.format_prompt(facts=fact_checker_output).text
    final_output = llm.invoke(formatted_answer)
    
    return {
        "assertion_output": assertion_output,
        "assumptions_output": assumptions_output,
        "fact_checker_output": fact_checker_output,
        "final_output": final_output,
    }

def extract_status_and_reasoning(final_output):
    llm = OpenAI(temperature=0.7, model=os.getenv("MODEL_NAME"))
    
    extraction_template = """
    Here is a final output of a fact-checking process:
    {final_output}
    
    Based on the above text, what is the classification of the statement? Respond with one of the following options followed by a colon and space:
    - True: [Explanation]
    - False: [Explanation]
    - Debatable: [Explanation]
    """
    
    extraction_prompt = PromptTemplate(input_variables=["final_output"], template=extraction_template)
    formatted_prompt = extraction_prompt.format_prompt(final_output=final_output).text
    extraction_output = llm.invoke(formatted_prompt).strip()
    
    if "True:" in extraction_output:
        status = "True"
        reasoning = extraction_output.split("True:", 1)[1].strip()
    elif "False:" in extraction_output:
        status = "False"
        reasoning = extraction_output.split("False:", 1)[1].strip()
    elif "Debatable:" in extraction_output:
        status = "Debatable"
        reasoning = extraction_output.split("Debatable:", 1)[1].strip()
    else:
        status = "Unknown"
        reasoning = extraction_output
    
    return status, reasoning

if __name__ == "__main__":
    if len(sys.argv) > 1:
        assertion = sys.argv[1]
    else:
        assertion = get_random_statement()
    
    print(assertion)
    submission = fact_check(assertion)
    
    # Print the detailed outputs to inspect their structure
    for key, value in submission.items():
        print(f"{key}: {value}")
    
    # Extract the final output for status determination and reasoning
    final_output = submission['final_output']
    status, reasoning = extract_status_and_reasoning(final_output)
    
    # Print the final status and reasoning
    print(final_output)
    print(f"Status: {status}")
    print(f"Reasoning: {reasoning}")

    # Record the result in MongoDB
    try:
        print("Attempting to insert record into MongoDB...")
        insert_record(
            script_name=__file__,
            script_code=html.escape(open(__file__).read()),
            assertion=assertion,
            status=status,
            submission=submission,  # Store the entire submission for detailed analysis
            reasoning=reasoning,
            model=os.getenv("MODEL_NAME")
        )
        print("Record inserted into MongoDB successfully.")
    except Exception as e:
        print(f"Failed to insert record into MongoDB: {e}")
    
    # Create a blog post on WordPress
    blog_title = f"Fact Check: {assertion}"
    blog_content = f"""
    <h1>Status: {status}</h1>
    <h2>Assertion</h2>
    <p>{assertion}</p>
    <h2>Reasoning</h2>
    <p>{reasoning}</p>
    <h3>Assumptions</h3>
    <p>{submission['assumptions_output']}</p>
    <h3>Fact Checker Output</h3>
    <p>{submission['fact_checker_output']}</p>
    <h4>Model Used</h4>
    <p>{os.getenv("MODEL_NAME")}</p>
    <h4>Script Name</h4>
    <p>fact_checker_mongodb.py</p>
    <h4>Script Code</h4>
    <pre>{html.escape(open(__file__).read())}</pre>
    """
    create_wordpress_post(blog_title, blog_content, status)