My Computers A.I page

This is the body of the webpage.



When it comes to intelligence we know we all have, all humans and even animals. But to think of a computer having intelligence used to be thought of as impossible but now we have "thinking" computers all over the web, including simple chat bots.
What is intelligence esentially, and what would we need to make a robot or computer have it. Intelligence is the ability to acquire and apply knowledge and skills. And one other thing to consider is the multiple intelligence theory. Howard gardner came up with the theory that there are esentially nine different types of intelligence, musical–rhythmic, visual–spatial, verbal–linguistic, logical–mathematical, bodily–kinesthetic, interpersonal, intrapersonal, naturalistic, and existential. And when we think of computers we think logical-mathmatical, not able to conceive the other eight.
Now there is one way to find out if a computer is truely, what we call, intelligent or thinking. The turing test, the Turing test is a test of a machine's ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, that of a human.
Today we went into making a chat bot. I programed my own and I am fairly proud of how it turned out. It asks multiple questions to you and I coded in multiple responses depending on what you answer. First I started off with this.

one part that was very difficult was how to get it to choose its own random birtday. I spent almost all morning trying to figure out how ot would randomly choose its own birth month and day.


I made multiple questions here and to get it to restart some areas was I put it into a loop so after you answer the last question it says, "Do you still think I am a robot?" you answer yes or no. And right after you can stop the loop by answering n or continue by pressing enter or typing in y.

I also made a "string" question that would ask multiple wuestions depeding on the answer you gave. For example it asked, What are you, then, what makes you human, if you answer emotions, it will give an answer based on that. As you can see below the intelligence one did go a bit farther making an answer for all nine possible answers to the one before it.

Today is the second day of AI and we were giving our bots a voice. Using a file steel, we imported espeak which when given the proper code would have the computer say what was after it. We made a few examples of it just saying hi, but I spent today remaking my chatbot from yesterday. Adding in voice to my text but leaving my text to see what the voice was saying.

Also looking online today I came across jaberwacky, a chat bot that you can set emotions on, and will even put in emoticons. The conversations are a little different than the ones with other chat bots, but they can get more interesting if you set emotions to like angry, sad, emotional, etc.

A problem i encountered was rooted from yesterday when I finally made it choose a random daay and month for its "Birthday". So i simply redid the code with the espeak so it would say it when I ran into the issue of them doing different dates. So it would say march 10th but would type december 8th. That is a big problem and So far have come up with no way to combat it.

on wednesday today we learned how to write into text files using Python. So we would type in the code and what we wanted to be typed into the file, the location of the file, and it would go into it. We also were able to see Mr farrel use his mike to speak words into the files. So what he spoke into the mike "not always accurate" would be typed into the file.

we also did a test that if we changed the text that was supposed to be written, if either it would add it to the doc, or erase and replace the text. As you can see below if you type in something new into the code spot, it will replace the text. So when we try to save something into file, multiple items, it could be an issue.

today we learned how to use python to search the internet using wolframalpha to find answers to our questions, or aka making our own factbots, as shown in the name, taking our question like a search engine it gives the best answer, or just the first it finds, and displays it onto the python shell. Changing it a little using what we learned yesterday we were also able to not only type the answer, but to say it aloud. One problem with that was is that it would also ask if you wanted to ask another question and this would overlay with the answer audio. One way we thought we could fix this was to simply add in a wait function, but because some answers were longer than others, that idea had to be scrapped, and I went looking for a solution to this problem. Using this we can have it talk to a chatbot and we would not have to program answers based on key words in questions. Having the code connect to the wolfram alpha that we downloaded let python connect to the wolframalpha search engine allowing us to search anything.


I figured out how to get the answer to the question you write put inside of the word doc we made before the somefile one. So now whenever your question is answered it is recorded on the word doc. One thing I have run into is that I know how to make multiple answers appear but at the moment, only the first questions answer appears on The doc, I am currently trying to figure out how to fix this.

Today we watched a video on the creation and design of WATSON, the smartest machine on earth. A long stretch from our little factbots that were basically search engines. We learned about computer learning that they put inside of watson, how he adapts from seeing other answers, documents, and old questions. When he analyzes old questions, because of the weird ways the questions are oriented, he finds patterns in the old questions to help him figure out what they really mean. Also he searches for key words in those documents that will be the keys to the answer, and searches his massive database for words and subjects that match up with all the clues he found And alter confidence in answers accordingly.
Today was our frst half day so we did not learn much, but make our own smart games. Starting off with the simple game of rock paper scissors. And moving on to a game of hangman which has proven a difficult challenge. To select words form a specific catagory randomly, and start the game. The difficult challenge I am facing now is matching the hangman position to if you got a lletter right or wrong.
today I finally finished my hangman game. After Three days of working both in class and at hime, I made a hangman game that works OK. There are a few tweaks that I will have to iron out but so far, It plays hangman like a real game of hangman. Defining the alphabet was a precaution I took to make sure that if someone put a number they would not get marked down a guess. One problem I still have not figured out how to fix is that sometimes, if you get the word and you are not hanged yet, instead of asking if you want to play again, it stays there and does not change, so you have to close and then restart the program to play again.

This is a quick little snip from my code that "outlines" my guesses and correct words with the hangman designs just above. This was one of the harder parts to make because I had to find some help online to make sure it all ran correctly.

Home