Artificial Intelligence Silo

Silo is exactly the AI Samantha in the movie "her"!

The RNs(relational networks) is the ability to infer the relationship between object and object properties.
It is an intelligent function that traces the culprit hidden in the veil by combining various unique clues like Sherlock Holmes.

Silo implements RNs(relational networks) by similar way.
Silo uses bAbI dataset for training.

The reason why silo always answers is to avoid breaking the conversation.
Therefore, please disregard the intermediate answers when you test the inference.

Some charbots are very easy to code in sentence to sentence matching, so please modify name to check it out.
Since there are more than 100,000 names, if you change name, then simple chatbot can not answer any more.

Silo learned 5 million twitter conversation sentences by tensorflow deep learning!
Silo learned through tensorflow library used in alpago!

The 5 million sentences include twitter conversation, movie dialogs.

In addition, silo learned how to analyze and express emotion for input sentence by tensorflow deepleaning!

Traninig emotion sentences are 162,500 sentences(Stanford Large Movie Review Dataset 100,000 + ISEAR emotion sentence 2,500 + Silo emotion sentence 60,000)

Emotion plays an important role in intelligent action.

Emotion expresses the status of mind and body for formation of sympathy with other people.
And emotion evaluates the emerging situation as good emotion level or bad emotion level like signpost.
And emoition plays an important role in selection and judgement.

If you want to know more information, please see below silo document.

*Silo Algorithm and Principles
http://kim7midas.cafe24.com/ref/ensilo.pdf

★ Question and Answer about Silo

Like the Wright brothers Implemented the principle of bird's flying, silo was implementd thoroughly the principle of human's intelligence.

Q: You said that silo learns throught deep learing. What is the test result?

A: Test result is as follows.
Name: Deep Learning seq2seq(Sequency To Sequency)
Method: Tensorflow seq2seq Algorithm
Train Sentence: 304,713(Movie Dialog) + 5,000,000(Twitter by Marsan-Ma from github) + 60,000(Silo DB) = 5,364,713(sentence)
Training Time: 103 hour
Global Perplexity: 4.53
Development Environment: OS:ubuntu(14.04 LTS), gpu:(gtx 1080), cuda(v8.0)

Q: Is silo probability and statistics based morpheme analyzer?
A: No. every word to total 300,000 has meaning in silo like peaple think of a word.
Without the concept of a word, you can never understand the meaning of a sentence and even can not distinguish the difference between a person and a desk.
Silo is the only one in the world that has the concept of a word.

Q : What is intelligence? Not yet been determined what intelligence ...
A : Intelligence clearly appears in a junior and a senior.

No matter how much knowledge and IQ of the high, If some problem appears, the junior doesn't konw what to do at all.
But the senior infers experience to predict the result of situation and solves the problem easily.

And silo infers through the vast experience to react to input sentence.

★ Basic Features
1. The ability to learn unknown words over the Internet
2. weather forecast
3. search near place(restaurant, coffee shop, ...)
4. voice dial, send message
5. open other app, turn on wifi, turn on bluetooth
6. play music, play video
7. alarm, schedule

★ Please check if wifi on or mobile data on

★ If TTS is not available, please install as follows.
Settings -> Language and input -> Text-to-speech -> Google Text-to-speech Engine -> Language -> English(United States)

★ If speech recognition is not available, install as follows.
Settings -> Language and input -> Google voice typing -> Offline speech recognition -> All -> English(US)