A chatbot – you digital external memory

Now and then we all do face a problem to memorize this or that, to remember to do something in time.

There are a lot of solutions: stickers, mobile applications, Google calendar, OK Google, which remembers a phrase after you have said “remember”, and various corporative solutions, such as : Evernote, etc.

All these solutions are quite practical, but they have some drawbacks.

The main one is: how to find the info stored in the memory?

  1. For example. “Remember: Buy tomatoes and chocolate in Walmart.” How to retrieve it from the memory? Where to write down to remember? How not to forget to remember to check what you need to buy.
  2. Or “Remember: home entry system code in San Francisco: 1,2,3,4.”

This info should be stored somewhere then retrieved. It’s important to remember what your need to retrieve and where to look for it.

Another problem is to remember info given to you by other people. Thus, my parents tell me the entry system code, reminding it many times via Whatsapp, email, sms… The most complicated thing is to find this info when I am at the door of their house. Before this moment I don’t even need to think about it.

Now it’s quite easy to do it via a chatbot!

The cloud memory solution functions as follows:

  1. You leave a message (or somebody else does it for you) in the chatbot.
  2. You ask it to remember: “Remember.”
  3. The entry system code in San Francisco: 1234.”
  4. The chatbot remembers not a phrase but the meaning. So if you ask it later, “What is the door password?” the chatbot will find the answer in its memory base.

Additional parameters for the memory:

  1. Who left the message (in brackets)
  2. The date of storing it in the memory (in brackets)
  3. Date and place to get access to the memory (in brackets)

More than one answer is possible, so the chatbot gives all of them by relevance as it remembers what it knows on the subject in succession.

Later on the user can erase the info from the chatbot’s memory if it’s of no importance anymore.

Thus, your chatbot keeps in its memory whatever you need and you can retrieve this info using any messenger: Gmail, Twitter, Messenger, Slack, Telegram.

You don’t need to think where all this knowledge is kept.

Current solution

A Russian company called Abbyy created a cloud service called findo.com that allows you to do a search inside your environment. For example, you may connect your google account, facebook, twitter and others and then find something.

But there is a big issues what do you have to do if you forgot completely what you tried to remember. Then this service is no good.

Create your external memory now based on a chatbot.

Memory options:

  • Thought, phrase, sentence, paragraph to remember.
  • Author of the information to remember.
  • Date of the information memorizing.
  • Date of the information expiration.
  • Memory key.

Usage of a public chatbot for one’s external memory:

  • Usage of a public chatbot without identification.
  • Registration (getting the PIN).
  • Access to your memory via your PIN.
  • Your external memory access opening to a second person via your PIN (you’ll have to either share your PIN or request the PIN of the person the access to whose memory you’d like to get )

So how does it work

We are using an API to create synonym list of meaning when you try to memorize something

API returns word embeddings for a given word or phrase in English or Russian.

  • Request:

    • http://paraphraser.ru/synvector?query=clean the room&top=5&format=json&lang=en

  • Response:

    • tidy the bathroom: 0.4533

    • tidy the kitchen: 0.4229

    • tidy the bedroom: 0.4076

    • tidy the upstairs: 0.3998

    • tidy the dining: 0.3753

So that means that we add synonyms and additional context to a memory slot.

This allows to add meaning inside the chatbot memory knowledge base and then a user can easily find what he is looking for.

For more info: http://endurancerobots.com/chatbots

Or email: gf@EnduranceRobots.com

+79162254302 (whatsapp, viber, telegram)