INTRODUCTION
A.L.I.C.E. (Artificial Linguistic Internet Computer Entity) is free natural language artificial intelligence chat robot. Intelligent Tutoring Systemsare programs that aim at providingpersonal instruction to students. In recent decades, conversational robots, also known aschatterbots, become very popular in the Internet and they are based on ALICE. TQ-Bot is a chatterbothelps the students during their learning process and supports theactivities of the teacher. This TQ-Botis able to analyse the requests made by the learners written natural language to provide adequate and domain specific answers relating the student to the right content . Besides, TQ-Bot is able to track and monitor the studentprogress by means of personalized questionnaires.
3.PEAS DESCRIPTION
Agent: TQ-Bot
Performance measure:1.Students score on course assessment test.
2. Scores based on the level of solution the user get from the agent.
Environment: Computers, Students.
Actuators: Screen display(exercises,suggestions,corrections).
Sensor:Logmap, M68332 microcontrollers,Keyboard.
4.TASK ENVIRONMENT SPECIFICATION
Fully/Partially observable: Partially observable
There is no surety that the students are equipped with the complete knowledge of the course after the tutorial. The agent may not always provide correct solution to the students questions. Hence it is Partially observable.
Deterministic/Stochastic: Stochastic
The next state of the tutor environment cannot be determined from the current state. Most of the partially observable agents are Stochastic.
Episodic/Sequential: Sequential
One stage of teaching process is connected with the next stage and the learning process happens sequentially. In an episodic environment the choice of action depends only on that episode itself. Hence the TQ-Bot is Sequential.
Static/Dynamic: Dynamic
The environment remains unchanged when an agent is deliberating. In TQ-Bot the students continuously questions the tutor. In a Static environment the agent doesnot not keep looking at the environment. Therefore it is Dynamic.
Discrete/Continous: Discrete
A discrete state environment has a finite number of distinct states. TQ-Bot also has finite number of stages hence it is Discete.
Single/Multi agent: Multi agent
An agent operating by itself in an environment is a single agent. But TQ-Bot
donotopertate by itself. Therefore it is a Multi agent.
5. PAPERS STUDIED
5.1.ANALYSIS OF PAPER 1
An intelligent web-based voice chat bot: A.L.I.C.E chat bot:
OBJECTIVE:
The objective of ALICE chatbot is to develop a artificial person whom it expresses its views as like as humans. It was also made successful by many chatbots especially ALICE chatbot.
A chatbot is an artificial person or other creature which involves in conversation with humans. This could be a text based conversation, a spoken conversation or even a non-verbal conversation. It can be interesting and intriguing. It appears everywhere, from old HTML pages to advanced social networking websites like Facebook,web,etc.. and from standard computers to smart mobile devices(iphones).
METHODOLOGY:
A.L.I.C.E. was written in the frame of Artificial Intelligence Markup Language (AIML), a good standard for creating any type of chatbot, also developed by Wallace. Most AIML interpreters are offered under a open source license. Therefore, many ALICE-BOT clones populate the internet, having created based on the original implementation of ALICE and its AIML knowledge base.
RESULT:
Though A.L.I.C.E was an interesting chatbot,it has many discomforts too.It may give same kind of responses or the default response, and it appears to take a long time to train a chatbot in another field of expertise.Most of the chatbots created this way are no longer accessible.
ANALYSIS OF PAPER 2
Imitation With ALICE:
OBJECTIVE:
Imitation is a powerful mechanism whereby knowledge may be transferred between agents (both biological and artificial).The major problems on the topic of imitation have emerged in various areas like artificial intelligence, including the cognitive science,robotics,human–computer interaction, embodied intelligence, software engineering, programming by example and machine learning.This paper describes a new general imitation mechanism called Action.Learning for Imitation via Correspondence between embodiments(ALICE) that specifically addresses the correspondence problem.This ALICE mechanism is implemented and its efficiency is illustrated on the“chess world†test bed that was created to study imitation from an agent-based perspective.
METHODOLOGY:
ALICE is used as a generic mechanism for building up correspondences based on any generating method for attempts at imitation by periodic assessment on the history of such attempts.Preliminary results using the ALICE mechanism implemented in the chess world. ALICE builds up functions as a kind of “refractive looking glass†or “prism†through which to transform a demonstrator’s behavior into the repertoire of the imitator’s own actions as constrained by its embodiment. Such a library of action correspondences can be employed when imitating .
The methodology used in imitating ALICE are as follows:
1) Displacement Performance Measure.
2)Sub goal Performance Measure.
RESULT:
Examining the different chess pieces as imitator agent embodiments established that the ALICE mechanism is useful in solving the correspondence problem when augmenting simple imitation sequence-generating methods. Depending on the specific affordances of the embodiment, the contribution of ALICE in improving the imitating performance can be either significant or only complementary.
6. COMPARATIVE STUDY:
ADVANTAGES:
• The main contribution of this technology is the use of an artificial intelligence entity
(AIML-based chatterbot) to enrich the learning process inside an e-learning platform.
This is done through both tutoring and evaluating students of a course using natural
language.
• A really important fact is that this bot can be easily integrated in open.source e-learning platforms such as Moodle and Claroline, offering a user friendlyinterface and the support of several languages.
• TQ-Bot is a tool that answers students’ requests (using natural language).It provides students a new way to look for information and guide them through
the resources of the platform while keeping a coherent dialogue.
• Feedback is allowed through assessing the provided answers, and so, further adaptation and improvements can be performed by teachers.
• This bot can also be used as a valuable supervising and evaluating tool. It allows students to observe their progress in the course and evaluates the quality of the questions provided. At last, teachers also can track student’s progress and change the adaption , the level of the questions and tests if necessary.
DISADVANTAGES:
• There is an “art” to writing AIML and creating a robot character, especially when writing the responses that have to given by the bot. This required really a lot of imagination and a sense of humor. It is bit difficult to learn AIML and give appropriate responses.
• The main drawback is that due to these technologies there is no human to human conversation is done.people sit for hours and spend their time on these aspects.So,The aim of creating this technology must be to build tools that help people, facilitates their work, and their interaction with computers using natural language; but not to replace the human role completely, or imitate human conversation perfectly.