Elephant and Monkey story in research| Dataknobs

Elephant and Monkey story

There were 2 villages. First village had an elephant. Second village had a monkey. Everyday elephants and a monkey used to find food in the jungle for their villagers.

Elephant used to go thru known paths in the Jungle. If an elephant found food, he could easily get it by the power of his trunk. Elephant could carry plenty for the whole village. If an elephant could not find food in a given direction, it was very difficult to try a second direction. Villagers have to sleep without eating food if the elephant does not find food in two attempts.

Villagers were unhappy with the Elephant as they could not get food everyday.

On another hand, the 2nd village used to get some food every day. Their Monkey was quick. Monkey used to climb and jump through trees. If Monkey did not find food in one direction, Monkey went in 20 different directions. Monkey could invent new paths. Monkey always found food. However Monkey was small. Monkey has to struggle to pluck food from trees. Monkey have to make multiple trips to carry food. Even after multiple trips the monkey could not carry enough for the whole village.

Villagers were unhappy with Monkey too as Monkey could not bring required food.

A tourist went to both villages and noticed both methods. He brought two villages together and asked Elephant and Monkey to collaborate.

Everyday in Morning, Monkey used to go in different directions and check where food was. Then Monkey used to sit on the Elephant back and navigate him to go to a place where food is. Elephant could pluck and carry plenty of food for both villages.

Both villagers were happy with this approach.

In above analogy

  • Elephant = Waterfall method, Monkey = Agile, Villagers = customers, Tourist = Data science Leader.

  • Elephant = Execution, Monkey = Research. Villagers = Customers, Tourist = Data science leader

  • We go through the same situation in Machine Learning and in research projects. We do not know which experiment will give the result. We should use a light weight agile approach in areas of ambiguity & research. (monkey approach)

  • We should resolve ambiguous items in the beginning. As we get confidence in experiment outcomes, we should use a more structured & heavy weight approach to get results. In the beginning of research it is hard to find how much food (ROI) a team will get, after the research team can draw conclusions on what ROI, Success criteria can be established for the project.

    Link on my Linkedin post - https://www.linkedin.com/pulse/monkey-elephant-story-manage-machine-learning-prashant-k-dhingra/

  • Blog 10

    From the blog

    Build Dataproducts

    How Dataknobs help in building data products

    Enterprises are most successful when they treat data like a product. It enable to use data in multiple use cases. However data product should be designed differently compared to software product.

    Be Data Centric and well governed

    Generative AI is one of approach to build data product

    Generative AI has enabled many transformative scenarios. We combine generative AI, AI, automation, web scraping, ingesting dataset to build new data products. We have expertise in generative AI, but for business benefit we define our goal to build data product in data centric manner.


    Generative AI slides

  • Unlock the Power of Language: Master Large Language Models with This Hands-On Guide
  • See best practices for prompt engineering
  • Evaluate whether you should use out of box foundation model, fne tune or use in-context learning
  • Most important - be aware of concerns, issues, challenges, risk of genAI and LLM
  • See vendor comparison - Azure, OpenAI, GCP, Bard, Anthropic. Review framework for cost computation for LLM