16 year old Farrell Eldrian Wu was named as one of the “10 Smartest Kids in the World” by Business Insider. He recently completed an internship at DataSeer where he helped out on our training courses. After the internship, we sat down with Farrell to talk about his passion for math and his experience as an intern at DataSeer.
DataSeer: You recently competed in the IMO. How did you do?
Farrell Eldrian Wu: At the 57th International Mathematical Olympiad last July 2016, I won the Philippines’s first gold medal (after its 28 participations). On a more detailed note, I was able to solve four full problems out of six, and made significant progress in one. One of the problems that I solved was a challenging combinatorial problem that was solved by only thirty-seven out of 616 participants around the world.
DataSeer: What’s your favorite area in Math?
Farrell Eldrian Wu: My favorite area in math is combinatorics, which is essentially a “none of the above” category within discrete mathematics that does not deal with the properties of integers. Within combinatorics, I am most interested in problems that concern algorithms (giving a step-by-step procedure to perform a process). I also enjoy combinatorial problems that deal with understanding the internal structure of a set-up and then observing its behavior as a variable becomes infinitely large or small.
DataSeer: Combinatorics huh? OK, a party is arranged for 14 people along two sides of a long table with 7 chairs on each side. 4 persons are best friends and wish to sit on the same side. Another 3 persons are best friends and wish to sit together on the other side. How many arrangements are possible?
Farrell Eldrian Wu: We multiply the following quantities (knowing that they are independent events)
One – There are 2 ways to choose which side would have the group of 4 and which side would have the group of 3.
Two – There are 24 ways to choose how the group of 4 will arrange themselves (4!), and 6 ways to choose how the group of 3 will arrange themselves (3!)
Three – There are 4 ways to choose which set of 4 consecutive seats (1234, 2345, 3456, or 4567) will the group of 4 occupy. Similarly, there are 5 ways for the set of 3 consecutive seats.
Four – There are 5040 (7!) ways for the seven students not in the groups of 3 or 4 to choose where to sit.
Multiplying, we get 29030400.
DataSeer: Wow, that was fast. OK, we believe that you won a gold medal at the IMO. If you end up doing research in math, what area would it be in?
Farrell Eldrian Wu: Although I naturally gravitate towards topics in discrete mathematics, I am open to research in any area within mathematics. Some topics within discrete mathematics that interest me are graph theory and algebraic number theory, which are both active research areas. Aside from mathematics, I am also considering research in computer science, particularly artificial intelligence, neural networks, and machine learning.
DataSeer: What are your thoughts on data science?
Farrell Eldrian Wu: It depends on the definition of data science used. Currently, data science is an evolving field, so different people and organizations would have different uses for the term. I believe that there are three components of data science: selecting the methodology, processing the data, and presenting the processed data, all on a large scale and for a practical purpose. The first two components are done on a small scale (small data set) in statistical research, while the third (without the first two components) is frequently done in journalism / news reporting. The second component, processing the data, is also done in statistics.
I feel that data science is not a completely new field, but rather, a combination of the powers of three different components. A data scientist must be strong at all three to be of help to his company. While it merely combines processes and strategies from three fields, it is the combination that gives data science its immense utility in today’s world, where data is more easily stored, collected,and analyzed with the use of computer technology.
The key word in data science is “science”. It is very easy for someone to process data in an unscientific manner, and the results would usually be inaccurate, causing controversy or even disastrous results as a result of the flawed analysis. There has been many reports of media using inaccurate data analyses that mislead the public. Making the process scientific, just as it was discussed in the DataSeer training, reduces the chance of an inaccurate or unhelpful analysis. Being scientific also makes it predictable and teachable to others – at the DataSeer training, even the aesthetic aspects were presented in a scientific manner, composed of established principles.
DataSeer: What do you suggest to improve our training curriculum?
Farrell Eldrian Wu: I think that the DataSeer training curriculum is already excellent. One suggestion I have would be to integrate a detailed evaluation system, similar to what I have done during my internship, and to allocate more time to assess each of the aspects of the group, instead of simply giving an overview of the group’s strengths and weaknesses. Another suggestion (but I’m not sure if there would be sufficient time for this) would be to allow students to practice each of the concepts covered, step by step, during the training. This may be done by presenting the group presentation task at the beginning of the day, and then interspersing the working time throughout the day at the end of each lecture, so that the learned material would be fresh when the groups work on the assignment.
DataSeer: Do you think an average high school student would be ready for a data science training? Why? Or why not?
Farrell Eldrian Wu: I do not think that the educational level of an individual would matter, as long as he is interested in analyzing and presenting data. I believe that what is more important is a strong quantitative intuition and skill and an analytical ability sufficient to identify trends. Although it may be true that mathematical and quantitative maturity comes with age, anyone who enjoys looking for patterns in the world, or reading analytical news articles such as those by “The Economist”, is likely to be ready for a data science training.
On whether or not a high school education (in the Philippines) would provide students an adequate background for a data science training, I would say that the Academic Track of the Senior High School curriculum develops the crucial critical thinking and analytical skills. For the mathematical / technical background, the “Introduction to Statistics” class gives all the necessary background for data science, as the main difficulty in data science is not technical, but rather, analytical. Core curriculum classes such as “Reading and Writing” and “Media and Information Literacy” also involves deep analytical thinking into passages or current events, rather than simply memorizing or understanding. Also, all students are required to undertake two research projects, which involve processing and reporting data (in response to a research problem), which is close to what is done in data science.
DataSeer: Thanks for your time Farrell! It was a pleasure having you intern with us at DataSeer. All the best in your studies in the US!
(Editor note: Farrell will be going to either MIT or Caltech on a full scholarship)