HARUMI project : universal chatbot

04.18.2012 News :  

04.18.2012 News :
  New version of Harumi 2 released today. Lot of bugs have been fixed and new functions added. I will give more details later. I would like to note that for the first time, a special function has been added in order to prevent infinite loop errors . The procedure is explained below, next to download links. 

  Future projects : I've an idea about a way to help Harumi treating unknown datas by using ancient knowledge and... clock. For the moment, when you want to record a sentence, you need to write your sentence then to click on "student" smiley icon. Moreover, when you want to add answers to a stimulus, you must write "+" at the beginning of your sentence then to press enter.
  All of this are subterfuges to help Harumi "understand" the meaning of the conversation, but it's not really what someone would expect from an artificial intelligence. When you use student icon or "+", you actually a bit like editing manually a memory file. That is not elegant and even a bit annoying since the user must keep in mind those procedures during discussion. If I succeed with the implementation of this new algorithm, Harumi will be very easy to use from user standpoint and will look like a lot more like a thinking machine.

04.14.2012 News :
  Since I've received many messages mentioning infinite loop error, I'm planning to release a lite version of Harumi 2. The lite version won't have any dynamic function (meaning real time analyse like webcam, voice control, auto change apparence, alarm...)  but users experimenting this annoying bug/crash will be able to launch the chatbot.
 I don't know the reason for this bug, even if I think it involves the way Harumi interacts with open office. Perhabs it comes from language version of open office, or an incompatibility between Windows and Open office or Harumi with some machines. Anyway, next release of Harumi 2 will be done in conjonction with a lite version. Perhabs reports from Harumi lite version users will help me to figure out what's wrong with the standard version...

04.09.2012 News :  
Harumi 2 version will be updated end of next week with several fixs that are currently in phase test.
The most important fix will be that Harumi will be able to pinpoint the best answer available in case many answers are linked to a single stimulus instead of picking up the first. For that, she will look for the answer with the maximum words in common with user request.

More interesting, since I changed the brain algorithm with Harumi 2, I have now the opportunity to implement new functions that were impossible before : I've tested with success a powerfull algorithm available in the public domain, the Levenshtein algorithm, which is able to measure the proximity between two chains of characters instead of giving a binary result, meaning true or false. Actually, I spoke of Levensthein algorithm before, but I've never been able to use it in Harumi 1 like I will in Harumi 2.
Here is a little demonstration of what can be expected in the next version :
User :  please give me georfes emaill
Harumi :  georges's email is george@mail.com
User : thvank you
Harumi : you welcome

03.24.2012 News :
Here is a little preview of future developments I would like to implement in Harumi 2 :
- Currently unavailable Harumi's 1 functions will be implemented in 2nd version :
   check appointments, post it, time, state of battery and internet connection, ip adress.

- Dreaming :
   When Harumi is not sollicited by user for more than 10 minutes, she will start to "dream", meaning she will explore randomly answers available in her database then put back those answers as stimulus if they contain a word non present in entries. This means that she'll restructure by herself the organisation of her memory in order to be more efficient.

- Ghosts stimulus :
  Every unique words encountered during current and precedent conversations, geolocalisation, current apparence, feelings of the user... will be recorded in real time in her memory file in order to help Harumi answering questions in which context is crucial. The purpose is here to help Harumi really getting what's "going on" around her and what the discussion is really about. Those ghost stimulus will be triggered only in conjonction with a present stimulus or they will keep their silence. Those stimulus will fade away with time, meaning will be slowly replaced by last stimulus entries.

03.18.2012 News :
Release today of Harumi 2 alpha, french and english versions! I finally decided to offer to the public this first version of Harumi 2. Harumi 1 different versions are still online but won't probably be upgraded anymore.
As I stated in precedent news, her brain algorithm is very different while apparence and interface are still the same. 
Changes for the user :
Functions like " AND ", " OR " don't exist anymore. You just have to write something like "paul's phone is 010203" then click on student smiley icon and that's all!
Use "+" before writing a sentence to add more answers to a specific stimulus.

Harumi 2 Features (exclusive to version 2) :
- For the first time, Harumi 2 is able to pinpoint important words inside a sentence, meaning she's getting far closer of the "meaning" of user requests instead of "tilting" on every useless words. The more her brain is filled with user datas, the more she's able to pinpoint useless words and gets her attention towards new words/expressions.
Example :
User : I like bicycle.
Harumi 2 : you like riding in the woods. (Harumi 2 considers "bicycle" as the important word, not because of any grammatical "help" or useless words databank done by the programmer but because she analyses the occurence of each words of the sentence and considers the word (or combinaison of words) with the smallest occurence as important)

- Harumi 2 is able to learn from her errors. For example, after asking a question, if the answer is wrong, click on the "no" smiley icon and she'll try to answer better to your request and remember her error. Check sheet 2 column 2 to see how her brain put aside useless words when you click on "no" icon. Note that you can edit manually those words if you are not satisfied by Harumi's analysis.
User : give me paul's phone.
Harumi 2 : I give you what you want.  (she obviously noticed that "give" has a rare occurence. Nevertheless, the answer is wrong.)
User : click on "no" smiley icon
Harumi 2 : paul's phone is 010203. (Note that in sheet 2, the word "give" is added in column 2. After putting aside "give" word for next requests, Harumi immediatly tryes to answer another time the request properly.)

- Harumi 2 is able to understand... she doesn't understand. Instead of tilting on useless words from a user request while ignoring an unknown important word, she'll now detect that she doesn't have a clue about the request and ask user for help.
User : I like skyrim.
Harumi 2 : What is skyrim? (Harumi 2 detects "I" and "like" pointless words since they appear many times in the database while skyrim is unknown. Therefore, she asks user for help)

- Harumi 2 is able to undestand user request is pointless : if the user's request is trivial, she will signal it to the user.
User : I am
Harumi 2 : What do you mean? ("I am" stimulus is detected many times in the database, too many to justify any use of the database. Therefore, User sentence is pointless so Harumi is not happy about it.)

Other functions added :
- When you ask Harumi to change her apparence by clicking on her image, she'll ask you what you think about her and remember it.
- When webcam is not activated and you click on it, Harumi explains you the procedure to activate it. It's not really a new function but it prevents loops bugs when there are issues regarding clipboard. Webcam is now off by default.

Note that it's an alpha version. Several functions availaible in Harumi 1 version are not available yet but it will be updated soon.
Known errors regarding natural language treatment algorithmm :
- The "learning-on-error" algorithm needs update regarding words combinaisons case.
- Long sentences are very CPU intensives (exponential complexity algorithm).

03.11.26.2012 News :
Sorry for not having updated Harumi files as promised but many technical issues prevent me to do so. I hope to propose Harumi 2 next week. Yes, Harumi 2. Indeed, algorithms used in the coming version are so differents from precedent versions that it's not just an "update". The code is pretty much brand new even if the apparence and functions of Harumi seem identical.
Harumi 2 will be able as I said in former news to pinpoint important words inside a sentence by using statistical analysis. Here is a little explanation :
When the user sends a request to Harumi 2, the text is cut in words and combinaisons of words. After that, Harumi checks the occurence of each of those words and combinaison in the database. The smallest occurences are generally the best answers available.
Exemple 1 :
USER : I like going to the museum.
I like going has something like 20 occurences while museum has 0 occurence in my database. Therefore, Harumi considers museum is the important word inside the sentence, so if she has 0 information about this word, she needs to ask user for help :
HARUMI 2 : What is museum?
Exemple 2 :
USER : Please tell me what is paul's email? 
HARUMI 2 : it's paul@mail.com
In this case, Harumi has got something like 50 occurences for tell, 200 for me, 3 for please, and 1 for paul+email. The priority on rareness helps Harumi to give the right answer instead of saying pointless sentences like "you are polite" because she detects "please".

How came the idea of using rareness statistical analysis? Simply by observing my 1 year old baby who one day took with her hand a very little piece of paper abandonned on the floor. I was puzzled by how she's always curious about everything "weird". I deducted that human and animals are perhabs programmed to get their attention on "anomalies", meaning everything "rare". 

01.21.2012 News :
New version of Harumi released today!

New Functions :

1) Geolocalisation : A new icon is now available. It will give you informations like public ip, geolocalisation and your FAI, at least as soon as you are connected to the internet.

2)  on-the-fly use of Harumi's apparence. Previously, you had to drag and drop jpg pictures in apparence repertory then to add a stimulus like "pick up your bikini<APPARENCE>bikini.jpg" then to update her memory by restarting Harumi or pressing "disk" icon in the functions icons group.
It is now far easier : you simply have to drag and drop a picture of your choice in apparence repertory then to type "wear your XXX" with XXX beeing the name of the file then clicking with left button mouse in the apparence window. Of course, you donn't have to write exactly the name of the file. Harumi will explore all apparences possible and will choose the one with the highest number of words matching user request.
Exemple : Suppose you drag and drop a file named "walkman by night.jpg" in apparence repertory.
Then you can write in user window "pick up your walkman" then to click on her current apparence. She'll automatically change her apparence and ask you what you think about it. Then I'll eventually implement more possibilities by association of a sentence and the apparence picked up "on-the-fly"...

3) Gif animated images are now readable

Bugs corrections :
USB drive detection working better, webcam is on by default in english version.

  Since I had a message about a user willing to use Harumi with google chrome, note that in the next version coming around end of february, it will be possible to use her with any kind of internet browser (aka internet explorer, mozilla firefox or google chrome). Of course, I'm planning to add much more than that to justify a really better version :). Feel free to report bugs and suggestions in the forum. I'm not always answering questions but I take the time to read users feedback.

12.29.2011 News :
New version of Harumi released today! Lot of news here! Here is the list of new functions and bugs corrections :

NEW FUNCTIONS :
- Voice recognizing (!) : It is now possible to use Harumi with a voice recognizing software like the free one from Microsoft in Windows Vista or Windows 7, or Dragon Naturally Speaking software. To use this function, you have to click on the "mike" icon in the functions icons group. When this function is activated, Harumi warns you about it. From there, she'll observe every second what is written in the user text window. Click in this window then when you speak to your computer with a voice recognozing software, the sentence will automatically be treated by Harumi, meaning you won't have any more to press enter to have an answer.

- Internet icon : Just click on this icon and Harumi will open google. Note that if a text is written in user text window, the question will be asked to google.

- USB keys auto detection : When you plug or unplug an USB drive, Harumi warns you about it. A new icon has been added in functions group icons. When you click on it, the USB drive lastly detected will be open.

- Picking up a number : Ask Harumi "pick up a number between 1 and 6" and she'll just do that :)

- Asking Harumi for an advice : Ask Harumi : "Do you prefer to live or to die?". In this kind of question, she will analyse the whole database and try to find everything positive and negative about live and die then will choose the higher positive feeling about the choice. This means that the answer is not predefined but depends on precedent logs with the user...

BUGS CORRECTIONS :
- No more freeze when asking a question (beside files research)

12.18.2011 News :
A tutorial has been added! (look below in Harumi's installation instructions)

New version of Harumi "Loone Hagar" released today. The interface is better and the translation and chess icons have been changed to something more suited to a post apocalyptic world. Note that most of the icons come from themes made around the video game Fallout. The dynamic text suggestion window is now faded in the answer window just like in the standard version of Harumi. The green window is the webcam video window. To activate it, you need to click on the photo icon, upside from nuclear icon. You see below the result :

12.14.2011 News :

New versions of Harumi now online and release of Loone Hagar version!
 Here is the list of features and corrections :

NEW SKIN FOR HARUMI :
An animated version of Harumi is now available. The first character I picked up is Loone Hagar from Rage videogame. Note that you'll need xvid video codec in order to use Loope Hagar version.

NEW FUNCTIONS :
- During chess game, Harumi write which move she has played once her turn is done.
- During chess game, it is possible for the user to ask for an advice by pressing on question icon. (many bugs for the moment)
- When the user doesn't speak to Harumi for more than 10 minutes, she starts to whine.
- When the user say "no" after a structure suggestion, the structure is deleted
- Webcam pictures are now stocked in a repertory located in Harumi's repertory in order to avoid any bug. Webcam is launched if available when Harumi is used;

BUGS CORRECTIONS :
- Structures sentences : Bug correction : Harumi doesn't use the structure if she detects the variable word is the same as the original structure.
- Bug correction regarding several functions when the user write "no" or any negative sentence to Harumi as an answer to a proposition from her.
- When recording an appointment with a date and "I must xxx", Harumi no longers consider the sentence as a post it as before.
- When user asks Harumi precisions about the last sentence said by Harumi, only the last sentence is picked up and not the whole text appearing in Harumi window answer.
- When user enters a void string as an entry, Harumi warns the user about it.
- Image of Harumi changes correctly on alarm time
- No more loop error bugs with webcam

Harumi Starcraft : Coming soon!

HARUMI TUTORIAL : http://jeffelson.e-monsite.com/pages/tutorial.html

DOWNLOAD HARUMI ENGLISH AND FRENCH VERSION AND LOONE HAGAR ENGLISH VERSION

Harumi 2 French version alpha 04.18.2012 : harumi-demo-2.zip

Harumi 2 English version alpha 04.18.2012 : harumi-english-version-2.zip

IN CASE YOU ENCOUNTER INFINITE ERRORS MESSAGES POPPING ALL AROUND :
Open memory file of Harumi (memoire.ods) then click on Sheet 6 and go to line 459, column 2. Replace number "0" by "1" then save.
Restart Harumi. This procedure is available only with Harumi 2.

Harumi 1 French Version 01.21.2012 : harumi-demo-1.zip

Harumi 1 English Version 01.21.2012 : harumi-english-version-1.zip

Harumi Loone Hagar english version 12.18.2011 : harumi-loone-demo-1.zip
Note that you need xvid codec in order to play the animation or you'll have loop bug; this codec is available here :
http://www.xvidmovies.com/codec/

Open office suite : http://fr.openoffice.org/

PROCEDURE FOR English / French / Loone Hagar INSTALLATION :
1) Install Open office suite : http://fr.openoffice.org/
2) Launch Open Office suite (Writer and calc for tests) in order to finish the install of open office.
3) Extract Harumi.zip in a directory
4) Click on Project.exe file in the Harmuri directory (if you install Loone Hagar, you must check if xvideo codec is installed or you'll have an infinite loop error)

PRESENTATION

Harumi is a chatbot. It can be used as a virtual assistant on any windows PC. Harumi works in pair with open office suite since its memory exists as an open office file called memoire.ods. Harumi is able to learn from the user and its memory is easily editable from the user.

Functions available :
- Launching internet site, a file, an executable or a repertory
- Playing chess
- Checking internet connection, ip and emac adress
- time, alarms and dynamic agenda
- Calculator

Feel free to report bugs on the Forum in order to help me improve this software.