Objectives
The goal of the Practice Module is to enable students to assimilate the knowledge and skills gained through taking the four Graduate Certificate component courses and apply this in a holistic manner to build intelligent software agents. The Practice Module consists of two parts: a Practice Project and an Examination. Successful completion of the Practice Module enables participants to demonstrate their proficiency across all of the skills that they have learned in the course modules and hence be certified as competent at the Certificate level.
Intended Audience
This Practice Module is targeted at participants who wish to complete the certification process for the Graduate Certificate in “Intelligent Software Agents”.
Prerequisites
Participants must have successfully obtained a competent score (or have been exempted from) the four component courses for the Graduate Certificate in Intelligent Software Agents as listed in the introduction to the Graduate Certificate page.
Components
There are two parts to the Practice Module:
- Practice Project: Participants will need to undertake one or more projects to gain practical experience and demonstrate their understanding and mastery of the skills taught in the four component courses. The practice project will require each participant to expend an estimated 10 days of effort. These days are not expected to be contiguous and may stretch over many weeks. These projects may be conducted by individual participants or in teams depending on the nature of the project requirements.
- Examination: Each participant is required to sit for an examination on a stipulated date and time.
- The overall grade for the participant will be based on the Practice project and Examination.
Typical examples of projects to be undertaken:
- Intelligent Software Agent: AirBnb Agent
Watch introduction video part 1 here & part 2 here.
With the travel industry estimated to be worth about US$11 trillion by 2025 there is a strong necessity for businesses to leverage on IT in order to claim a piece of the market and stay competitive. At the same time, consumers are spoilt for choice and always on the lookout for the best deals. Accommodation planning is a big part of any travel plans and consumers may not always have the time to source for places to stay as well as building an itinerary for their trip.
To address this issue, the team has decided to produce a solution which will act as a personal concierge for users and look for best recommendations for accommodations and it's nearby top recommended venues such as restaurants and tourist attractions. Our personal concierge, also known as 'S.A.R.A' which stands for 'Search And Retrieve App', is a chatbot which will leverage on natural language processing and robotic process automation techniques to assist users in their travel planning activities.
Chatbots also known as “conversational agents” are basically software applications that mimic written or spoken human speech for the purposes of simulating a conversation or interaction with a real person. Currently there are two primary ways chatbots are offered to users which are web-based applications or standalone apps.
- Virtual Process Automation Agent: Intelligent Job Hunter & Cover Letter Writer
Watch introduction video here.
Not only students in NUS-ISS, almost everyone has experienced the job hunting, especially near the period of graduation. According to the survey, as of 2016, the number of recent graduates in Singapore has reached 8 million. But it is undeniable that the entire job search process is very cumbersome, the process of scanning through multiple job postings to find relevant ones is time-consuming enough, let alone to write cover letters manually respectively for different job positions.
Although there are many cover letter generators already, but on the one hand, they only focus on the format of the resume and cover letter without content which need be typed in by customers own, on the other hand, they ignore the process of selection jobs before cover letter generation things and the process of job application after generations. Our team has built an agent which can help us do the whole working flow of finding a job.
The system is the Intelligent Job Hunter (IJH). After the user registers with our system and uploads their personal information and existing resumes, our system will firstly extract the information from the existing resumes and save them into database for the comparison in the next stage. Every two hours in the morning, our system will automatically collect all of the job positions through email, compare those job positions to the information extracted before and select out those job positions that are suitable for the user and generate cover letters for all suitable positions respectively. After these processes are completed, our system will send the selected job positions to user through social media messenger, user can not only just select the job they want to apply in chat window, but also can go to our website to check the cover letter generated and modify them. Then after user confirmed, our system will send emails to the hiring manager automatically.
To build the Intelligent Job Hunter, we designed a hybrid solution combining the web application and chatbot, using the Vue.js, Flask, WeChat Subscription, TagUI and Natural Language Process.
In this project, our team applied the knowledge and techniques acquired from the lectures and workshops into the reality business application scenario, and we would recommend our Intelligent Job Hunter to every student in our institute, hoping it can help them improve the job finding efficiency and quality, help them to find a better job
- Food and Workout Recommendation Intelligent Self-Learning System
In today’s fast-paced society, it is very challenging to eat right and exercise right. Most people know the benefits and desire to eat a balanced diet and to do the right amount of exercise, but many just do not have the time to plan what to eat and how much and what exercise to do. Some tried to count calories but there is up to now, no quick and accurate way to do so.
On the other hand, eating is also an activity which can bring happiness. We all feel happy when we are served our favourite food. How do we maximise this happiness and at the same time avoid putting on extra weight which makes us feel guilty the next time we ordered that chocolate ice-cream?
The proposed solution is an Intelligent Self-Learning System which recommends to the user a highly personalised daily eating and exercise plan which keeps the calories in check and maximises the user’s happiness from food consumption. The exercise plan is also being optimised such that it is just enough to burn away the input calories by an amount based on user’s weight loss goal.
The system was implemented using various reinforcement learning and evolutionary learning techniques, namely: Q-Learning, Genetic Algorithm, Evolutionary Programming, Evolution Strategies, Particle Swarm Optimization and Differential Evolution. When live data from IoT Sensors and user feedback is available and is integrated in a production system, the system will be capable of continuous self-learning and self-evolving according to the changing environment such as changing user food preference, new food items and new exercises.
- More example projects in other domains can be found here.