2017: Etsuko Ishii

 Etsuko Ishii
Home University: the University of Tokyo
Field of Study: Applied Mathematics
Status: B3   Expected Graduation: March 2019
Research Host Lab: Prof. Devika Subramanian, Department of Computer Science
Research Project: “Analyzing Houston Flooding Using Unsupervised Machine Learning.”

Why Nakatani RIES?

Science & Engineering Research at Rice will affect both my short-range and long-term visions. In the short-term, it will encourage me to study abroad for several years. To realize my dream to become a researcher, I plan to join in a graduate school in the U.S. after I get a bachelor’s degree from the University of Tokyo. I took an intensive German course in Free University of Berlin last winter, which inspired a desire to study overseas again. I expect this participation at Rice will serve as a good simulation for it.

In the long-term, it will be the first step of my career as a professional in science and engineering. I hope to carefully observe during this internship program what researchers do in their studies. I know how they look like while working, since my parents are linguistic professors. They often brought me to conferences on linguistic or a comparative literature to show me the academic world since my childhood. However, there is a serious gap between just seeing others and trying things myself. I heard that Rice University takes good care of students, and in addition, plenty of attractive research is conducted there. Therefore, it will be appropriate for my first academic research experience.

Goals for the Summer

  • Get used to academic English.
  • Understand how scientists conduct their research.
  • Broaden my sight through communication with people from various backgrounds​.

Excerpts from Etsuko’s Weekly Reports

Week 01: Arrival in the U.S.

“Where are you coming from, Mexico?”

I was a little upset when an immigration officer asked me this a question, after showing my passport and DS-2019. This was the first time, and surely the last time in my whole life to be mistaken as a Mexican. After my answer, absolutely not, we gave a wry smile for a while. He was simply too tired. It suddenly reminded me of the current tense political condition and the location where I was standing.

Great differences among countries always appears in such trivial things. Just one sentence, just one custom has power to defeat us entirely. We cannot ignore cultural backgrounds, as Sarah told us in the orientation at Rice, but we can accept them.

As for me, I experienced many aspects cultural diversity during this first week; and all of them wer acceptable so far. Everyone here expects that everyone should speak English, as I expected, and this seems very “American.” Therefore, I was surprised to see Spanish is also written on some signs. The campus-tour guide told that Rice University is the most friendly campus in th most friendly city, Houston, in the most friendly state in the U.S., Texas which I felt was true. People are willing to help each other, so I can always ask them to show me the way. This is a stark contrast to my university, the craziest campus in Japan. When foreign people talk to me at my campus, they are mad or solicitors!

I wanted to use this opportunity to think about my future career. I acquired a lot of helpful information about careers for PhD students during the pre-orientation in Tokyo and from Prof. Kono. The statistical data he gave us in his lecture, for example average terms to gain PhD degree, were things I had wanted to know more about.

I spent most time of my time this weekend shopping. I visited H-mart, a Korean supermarket, on Saturday, and the Galleria, the biggest shopping mall in Houston, on Sunday. I was overwhelmed by spaciousness of the buildings. I got a lot of nice foods and clothes! I bought a book this week: They Say, I Say. Through this book, written by Graff and Birkenstein, I can learn how to describe matters objectively.

Preparing for Research in the U.S.

With Prof. Subramanian and Prof. Kono at the Welcome Reception ~ Etsuko Ishii

Before I arrived in Houston I exchanged a few e-mails with Prof. Devika Subramanian and my mentors. Prof. Subramanian’s research interests are in the areas of artificial intelligence and machine learning and their applications. She told me that I would work with hurricane flood risk assessment, so I read some papers that she recently wrote on hurricane wind risk. I also tried to read a textbook about deep learning which they recommended. I bought other basic books in Japanese on machine learning, because that textbook seemed a little difficult for me.

While I will be engaging in research on hurricane risks, Prof. Subramanian has other projects, for example modeling in medical contexts. This flexibility across scientific fields represents an aspect of computer science as a third tool in scientific research: experiment, observation, and numerical analysis/computational modeling. I could not be aware of necessity to understand variables to conduct computational modeling until I began this research internship. I expected to work here more with machine learning and python, but now I am working on a tutorial how to deal with geographical data. Once modeling methods are established, we can apply them to various problems. We can invent an all-around instruments in a sense. This is the most attractive side of computer science, which I discovered during this week.

Suggestions for Future Fellows
Acquire the main ideas of what your professors or mentors give you to read before your arrival. They never expect you to grasp everything, especially the details. It is most important to distinguish what you could understand and what you could not. Then you can have a constructive discussion with professors and mentors.

Question of the week

WHY are you still using the yard-pound system, even in the scientific field?

Return to Top

Week 02: First Week in My Research Lab at Rice

Stocking up on groceries before Harvey ~ Etsuko Ishii

Hurricane Harvey was a major concern in Texas this week; also, hurricanes are the main research focus for me this summer. People are fed up with being worried only about Harvey all-day. However, I’ll spend my entire summer to consider hurricanes. I even felt lucky to suffer this disaster, otherwise I have been pretty stressed lately.

The first day in my lab was not as special as you may expect. My safety training took only 10 minutes. Instead, I talked about the overview of this research project with Prof. Subramanian. On another day, I had a time to introduce myself to my mentors, Isaac and Joe. They spared me as much time as possible for me, although they are noticeably busy with their research too. As is often the case of computer science labs, we work when and wherever we want. The special advantage of this style is that I do not have to stop my research even during the hurricane. I exchanged information mainly via email, so my ability to write emails in English improved rapidly.

I am now learning how to use ArcGIS, the mapping and analyzing platform, at GIS/Data center on the basement floor of Fondren Library. I am fascinated by the multiplicity of this tool. During her session on developing a research poster, Dr. Gayle Moran repeated the importance of attracting attention to your own poster. My capability will be tested as many people have a keen interest in my research theme.

Trying out a veggie burger ~ Etsuko Ishii

On Tuesday evening I went to the Rice Farmers Market and bought some bread. There was a stand which was selling “Bones to go.” I wonder what it was! Unfortunately, I cannot check them this week. Is that a kind of “Tonkotsu(豚骨)”? I also tried “vege-meat”, whichi is something I wanted to try during this program. The hospital where I was born serves vegetarian food to all patients. This was my first time trying vegetarian food. Since then, I myself never want to be one, but have continuous interest in vegetarian food or vegan food. I enjoy cooking and wish to adopt them as one of the options. Before I came to the US, I watched a documentary program about alternative food, especially focused on alternative protein. Then my desire to try “beyond meat” grew up! It looked and smelled the same as normal meat, but its texture was a little odd. I will buy another product next week.

Question of the week
Vegetables here are very similar to Japanese ones in appearance, but they do not react in the same way as I expect. Is it because of the difference in varieties or soil?

Rice Farmer’s Market ~ Etsuko Ishii

Research Internship Update
My research goal is to predict flooding risks in Houston at street-level, given rain fall data. Civil engineers have already developed several models to predict risks with a high degree of accuracy. However, they need plenty of time for calculations; often too much time to get ready for coming hurricanes. Therefore Prof. Subramanian’s group has applied machine learning methods to this field. If we can predict with 75% accuracy before hurricanes arrive in Houston, it is more practical and helpful.

I combined the following data in 2011-2015 to construct a model, using ArcGIS: 311 calls of flooding cases, rain falls, digital elevation model(DEM). Then I pass variables to Python machine leaning system.

This week I learned how to use ArcGIS and the model my mentor Isaac developed for the wind risk prediction. I also need to get accustomed to Python and machine learning. I made a heatmap by plotting all 311 flooding calls in 5 years. If you look at this heatmap, you can easily understand which sections of Houston are likely to flood. I knew that the crossing in front of the hotel flooded several times in 5 years. While I was dealing with 311 data, it flooded again! Empirical knowledge was beginning of the modern science. Machine learning will become an alternative “experiment” method in the future.

Return to Top

Week 03: Interview with a U.S. Researcher

I interviewed Zhouhan who is in a master’s program and who has working in the Subramanian Lab for three years. He was born and grew up in China, graduated from a middle school in Shandong, and a high school in Tianjin. Then, he moved to Houston and entered Rice University as a freshman undergraduate. His first interest in computer science was aroused by a class he took when he was a freshman. Creating a small game in Python each week as an assignment, he was attracted to the ability to create completely new things by himself, so he decided to major in computer science.

He is now doing machine learning research on Twitter. He recently published a paper about Twitter bot detection, which he will have a present this weekend in New York. He developed a Twitter spam detector which can identify and visualize 10,000 spam accounts every day. For this project, he usually cooperates with political scientists with his projects. Therefore, other projects he does are more political; he completed an analysis of Twitter censorship of government in Turkey and a sentiment analysis of political events, Hong Kong Umbrella Revolution in 2014. While establishing a model to analyze tweet data, social scientists tried to describe the political background or changes in public opinion.

He is thinking about working as a data scientist in some large company in the future, because it is usually easier to get an access to big data for analysis. He already participated in two internships in industry. One was at APP Appie, which is known for providing data and tools for app analysis, where he joined in a data warehouse team. The other company was Quantlab Financial, which is a group of scientists and engineers doing research on finance using their knowledge of scientific fields. There he became a member of engineering team to build a model to predict market motions in high frequency trading.

In addition to the interview above, we talked about our own cultures. As we have several common features in our culture, like classroom rules, this was a good opportunity to point out differences between US and Japan/China. He has been to Tokyo and told me some experiences during his stay, which I found very interesting. We exchanged opinions about public transportation in cities. Moreover, we shared the political opinion trends of younger generation in each country. His motivation of Twitter censorship or depression comes from what Chinese government does to Chinese people. He also wants to travel to North Korea to see the present status there.

During his interview, I found that as a data scientist, or other computer scientist in applied projects, it looks more attractive to work in industry; more salary, more data, same working hours, same project. I would like to participate in some research internship programs in industry, but in Japan I rarely find open positions. I think Japanese companies should open their internship positions not only for recruiting new graduates. Anyway, he gave me some tips for my future career beside working. One of them is that I can satisfy my curiosity in different fields at the same time. I decided to enter my department, mathematical engineering and information physics, because the students there told me that I can do anything I want to learning how to use applied and computational mathematics. I can now give the same advice for younger student in the same way as other students gave me. Just join computer science, or mathematics, then discover what is most interesting to you.

Interview with a U.S. Undergraduate Student
I also interviewed with one of my mentors, Isaac. He is a senior undergraduate majoring in computer science. Although he has never participated in international research or exchange programs, he has had many experience in international environments for two reasons.. First, Rice University is such a diverse community that it is often the case that he is only student who was born and grew up in US. He has already spent more than 2 years with Prof. Subramanian, who comes from India, and in her lab there are more students from Asia than the rest of the world. Second, computer science projects allow people to be involved regardless of the place they stand. With technology these days, most of them originated through the development of computer science itself, the physical distance is no longer such an obstacle.

In addition, what I was most interested in when talking with him is his internship experience at UBER this summer. He mentioned several differences between working in industry and in academia in computer science. First of all, the amount of money you can get industry. It is not so unusual that computer science students can earn enough money through a summer internship for both their school costs and have some left over for personal costs; much more than in academic research. Second, it is possible for them to keep publishing a paper every year, especially in the machine learning field. Researchers of major IT companies, i.e. Google, Apple, IBM, present their work every year.

This opportunity was helpful for me to consider my future career, though my major is not computer science but applied mathematics. I think the differences he suggested are also seen in my field. I would like to participate in some internship programs in industry after I enter graduate school to discover which environment is more suitable for me to work in.

Research Project Update

Sushi ~ Etsuko Ishii

Although I can work on my research remotely, I could not make much progress on my research this week, due to Tropical Storm Harvey. Prof. Subramanian’s house was damaged seriously and Rice campus itself closed to allow faculty and staff to spend time this week for recovery. Instead, I could participate in various fun activities this week. I spent a lot of time to dancing and the Japanese Fellows even went to a two-step dance party. Savannah, who was US fellow this summer, invited us to this party held at her residential college (dormitory). I also joined a ZUMBA class at the Rice Recreation Center on Tuesday evening, which stayed open during the week campus was close. I could not follow the motions of the instructor, but it was good workout, as I had to stay still at hotel for a week.

At the Japanese Language Table Discussion Session ~ Etsuko Ishii

We also got to attend a Japanese lunch table discussion session with Ozaki-sensei where I met many Rice students who are learning Japanese. I had dinner with them twice, once at a hamburger restaurant in Rice Village, another time at Kula-Revolving Sushi Restaurant currently opened in Houston near Chinatown. I never imagined that my first Kula-Sushi experience would be in outside of Japan. It tasted almost like real Japanese sushi while there were also are various types of California rolls. I also had an opportunity to have lunch at one of the Rice residential college cafeteriaa. One of the students I met at Japanese lunch table took me there, after I happened to meet her at the Rice Rec Center while I was swimming.

Swimming at the Rice Recreation Center ~ Etsuko Ishii

I also began to work my poster this week, decided the overview of the poster design, and now writing the introduction and methods part. I am now trying to combine geometrical data and 311 flooding data to see whether the location of flood is the lowest point compared with surrounded area.

I am using a textbook Deep Learning written by Goodfellow et al., and its Japanese translated version PDF that was uploaded by Prof. Matsuo Lab in the University of Tokyo until it is published. I felt lucky to be able to begin reading the translated version, of course helping them by pointing out the mistranslated sentences or misconversions. In addition, I found a class dealing with machine learning that will be offered next semester in the Faculty of Economy. I can continue to study machine learning after I go back to Japan!

Question of this week
Why are there so many blocks of dry grass on the road? I think they are typically seen in desert areas, while Houston is humid.

  • Yes, Houston is quite humid but the sun is also very hot and this heats up the roads/pavement. So, if the grass along the roadway is not regularly watered the reflected heat from the roadways can still cause it to burn/die and turn brown.

Return to Top

Week 04: Reflections on English Language & Life in the U.S.

People in Houston seem to be very talkative. There are various situations in daily life where people often have a brief conversation (small talk), i.e. waiting at a street light, taking an elevator. I also engage in short conversations with them; they are the most helpful chances for me to practice speaking in English. I try to use the phrases that they use to me in the same situation, like conversation exercises, A-B-A-B then B-A-B-A, although the partner I practice with is not the same person.

The biggest problem I have in speaking in English is confusion of English and German. My language background is a little odd. I am aware that I sometimes use strange English. First, my accent is not a typical Japanese accent. This was pointed out during one of the webinar by Prof. Shuman, that I have a western-European accent. This was not first time it was noted that I have an accent, or even a German accent. There are many books or lectures in Japan to correct Japanese accents, which are powerless for my problem. I looked for pronunciation books in Germany last summer, but found nothing worth reading. I came to a conclusion: if others can catch me, it does not matter whether I have an accent or not. Therefore, I make most of this stay in Houston to find out the way other people can follow me.

This weekend we got to visit NASA Space Center Houston ~ Estuko Ishii

Second, which is more serious, is grammatical and vocabulary problems. I am too influenced by German expressions or grammar. I finally learned to answer “Yes” or “No” this week; otherwise I unconsciously answer with “Yeah (Ja)” or “Nee (Nö)”. The reason why I needed so much time to acquire such an easy reply is that people can understand what I mean, even if I used the wrong one. It is also hard for me to use simple present tense, not perfect present tense in conversation, especially in interrogate sentences. I often confuse word order when I speak a long sentence.

My strength in handling incomplete languages is that I know how to make others understand me and how to grasp others are saying. However, I found that this “others” should be native speakers of that language. When I had a discussion with my mentor Zhouhan, whose mother language is Chinese, my method did not work. We were fortunately able to solve the situation by showing our ideas in Chinese characters.

At NASA ~ Etsuko Ishii

By the way, I can apply one technique of this method to any language. Imagine you have to use a word that someone tells you, but unfortunately you cannot catch it, although the conversation partner has repeated already three times. Just remember first two or three sounds and last one sound, and its intonation or accent. When you pronounce it, emphasize the beginning and the end sound, and leave the middle ambiguous. Important thing is trying to reproduce the intonation and accent as natural as possible. You need a little practice, but I think it works in 80% of those situations.

Research Update

Finally, I debugged a program my mentor gave me to handle lidar digital surface model(DSM). DSM is a digital 3D model of terrain surface that contains buildings. I began to study machine learning systems, in which I am highly interested. I will use the same machine algorithm that my mentor already developed for wind risk assessment, but I need to preprocess to input variables. I am now trying to find the best way of pre-processing.

The algorithm I will use is application of an image recognition system called VGG16 convolutional layers. VGG is a name of the team which developed this method. Convolution neural network(CNN) is the one of the most popular approaches in machine learning. It is mainly used in image processing fields because it was studied to recognize hand-written zip code on letters at first. Facebook, LINE, Google, Spotify and more famous companies are using CNN for their services; tagging by identifying faces in the picture, recommend suggestions that the user may also like, voice retrieval.

I will use this VGG16 convolutional layers as a feature detector of DSM images. More precisely, I will combine a part of VGG16 convolutional layers with trainable layers. VGG16 can extract general features of images and trainable layers can learn best function values for flood risk assessment. This method is called fine tuning or transfer learning. It derives from a curious phenomenon of deep learning seen in common. Features learned on the first layer tend to be similar to Gabor filters and color blobs. Moreover, features of the first layer do not depend on specific data sets or models and can be applied to many situations. We can save a lot of cost on calculation by using fine tuning.

The most difficult and important thing to build a machine learning model is to choose the most preferable features as inputs. These past two days I have been struggling with this problem. The approach structure itself is more general than specific. What makes a difference, and the best model for a particular task, is from which aspect are you looking at your task.

Question of the Week
In Rice campus, I can enjoy many artworks. Is it usual to exhibit various works of art inside university campus, or is it a feature of Rice Campus?

  • This varies by school but yes, on many university campuses in the U.S. there are public art installations or exhibits in their art galleries or departments.  When David Leebron, our current president first came to Rice he made it a goal to expand the number of art installations on campus so in the past 10 – 12 years there have been may new pieces added. For more, see the Rice Public Art page.

Return to Top

Week 05: Research in the U.S. vs. Research in Japan

Working with Devika at lab ~ Etsuko Ishii

It may sound odd to you, but I am still not sure how it is different to conduct research in the U.S. compared with in Japan. There are some clear contrasts that everybody is aware of. People discuss things more frankly, and even some professors like to be called their first name. I also called my host professor Devika, not Professor Subramanian.

However, the lab culture depends much more on what field it is in, or who the professor is. We computer scientists can do experiments in our bedroom, which is also true in applied mathematics. My departmental labs in Japan have some mandatory meetings several times a week, but at other times you can spend your working time as you like. This is not special thing in our field. We share this culture worldwide. Devika encouraged me to run my own project, even a visiting undergraduate student like me. I do not know whether it is often seen in other labs in Department of Computer Science at Rice, or in the U.S.

What I noticed during these five weeks is that your experience in your research will be highly influenced by your lab. All Nakatani fellows were doing research at Rice University, but no one had the exact same experience. Before this program, I did not understand the importance to take into account who the professor of the lab is tht I want to join in. I was just curious about what the lab does. Lab culture and people are an outcome of who the professor is.

I did not like terribly sunny days in Houston. I did not like roads flooding easily. I did not like squirrels kicking my feet to steal my lunch. But I really love Houston, and really love Rice University. I felt comfortable working with the people at Rice and talking with other undergraduate students. I will miss people at Rice and the friendly people in Houston.

My summer vacation here was much more than I expected in Tokyo. I thought I would follow what my mentor said, or do a small part of big project. How lucky I am! Thank you all in Houston who supported me these weeks.

Final Week in Lab
This week was incredibly busy. I kept working until midnight the whole week. There was no farewell dinner/party in my lab. This does not mean they are avoiding keeping in touch with me, nor are they too busy to spend time for me. Rather, I decided to continue to work with them after I go back to Japan. My research began to go well, and what I could do in this 5 weeks was just a first step of our goal. So, there will be no farewell since I will continue to be a remote member of the lab.

The first half of this week was challenging. I collected all the data I used as inputs and converted them into suitable formats. In the methods part, this process is usually described only by one sentence. In some cases, there is no description at all. However, this is the hardest, most important, most time-consuming task but also the most boring process in establishing a new simulation model. I had to make many compromises because of time restrictions. I struggled with huge size of the data, slow-speed of my Wi-Fi line, incorrect data and deadlines.

The latter half was exciting, while I was confused by not having enough time for everything. The analysis part is a true computer scientific experiment. It is fun because we can get some results. I was frustrated that I could not even start my experimental part until Friday. Anyway, I spent the final weekend trying to make progress in my research, and it was worth doing so.

I really enjoyed the poster presentation session that was held in Duncan Hall at Rice University. I could not practice in advance as much as the other Nakatani Japanese fellows did. They prepared very well. I really appreciated what they did during this past week. There were two reasons why I could still be confident during the poster presentation session without practice. First, I know everything about my project. I am the person who lead this research. Devika and my mentors helped me a lot, but I dealt with almost all parts. Second, instead of practicing, or a writing script for presentation session, I talked about my project to other people with various backgrounds the day before the presentation. That worked well. I learned how to talk about my research.

I also have a mentee now. She is a sophomore student in the Department of Computer Science, interested in machine learning solution of flooding. She will help me with my project, especially collecting the data part, since when I’m in Japan my access to information of Houston is not enough. I will also help her to learn mathematics behind machine learning.

I’ll do my best to complete my project. I’m looking forward to showing you my final results soon!

Question of the Week
Why do I have to go back to Japan now!?

Return to Top

Week 06: Final Week at Rice & Research Poster Presentation

Poster Presentation

Etsuko Ishii at the poster session.

I really enjoyed the poster presentation session. This experience was completely different from the presentations I have done before. This was more like a conversation. I simply loved everything about the poster presentation activity. I liked talking with people, I liked sharing my thoughts, I liked to hear ideas from other people. I could even use grammatically incorrect or informal words. The opportunity I had the day before was also helpful. I modified the logics for each person to grasp my research overview depending on the background of the person. Because of Harvey, many people were interested in my research field.

Final Research Project Update

With Devika at the poster session. ~ Etsuko Ishii

My host professor was Prof. Devika Subramanian in the Department of Computer Science. I had two mentors, Isaac and Zhouhan, but I worked mainly with Devika, not them. I fortunately could have my own project and do everything by myself. My mentors gave me suggestions or advice whenever I asked them to.

My poster title was “Analyzing Houston Flooding Using Unsupervised Machine Learning.” My research aims to establish a flood simulation and/or prediction model for Houston. Tropical Storm Harvey was not the only flood Houston has experienced in the past 5 years. Some regions of the metropolitan area have flooded more than five times in 2011-2015; in the worst places up to 14 times. There is a “floodplain,” with a flood risk hazard map defined by FEMA. However, its accuracy turned out to be only roughly 50%. 51.2% of homes flooded in 2013-2015 experienced more than one flood, even though according to the map it was estimated that they would never flood in 500 years.

First, I tried to extract the flooding cases where there were multiple flooding events from 2013-2015; although they were not including in the FEMA floodplain. I used an unsupervised machine learning method, called “k-means clustering.” Clustering is a way to group objects into some classes based on its “similarity.” Using k-means clustering, we can observe the input data structure, or the relationship between the features of the inputs. After I extracted those groups, I made elevation models of each point in the group (about 2500 points) to see what was happening there in terms of its geographical aspects. Then I tried another k-means clustering to classify them. I successfully divided them into four groups.

The region with multiple flooding events are characterized by high density development and proximity to large tracts of land at low elevations. Further analysis of these homes at different scales (50mx50m to 500mx500m) is needed to identify key factors that cause flooding, as it is not so clear with LIDAR DSM (Digital Surface Model), what I used as a geographical feature, alone.

After I identify the key factors of flooding, I will build a fast and accurate flood prediction model at the home level using diverse data sets rather than by just running detailed physical simulations. I’ll get more a precise floodplain than FEMA did!

Return to Top

Week 07: Visit to Lehigh University and Return to Japan

I felt as if it had been not the first time to be there in Philadelphia and Bethlehem. Everything looked familiar to me and I was really relaxed. At the same time, I had a strange feeling; everything is too new. I’m sorry if this sounds rude, but the history of the town is not so long as that of in Europe. I want to visit again because I was too tired to look around. The lectures at Lehigh University were interesting, especially the sessions on our future. The personality test analysis was helpful for giving me insights about myself to consider when applying to other internship program abroad. The resume template will also help me.

The working environment at Sysmex was wonderful. I want to work in a company which has such ideal circumstances to conduct a research. The final presentation session was a great opportunity to hear what other Japanese fellows did during this summer in JAPANESE. It sometime happens that ‘band gap’ was the only word I could understand when their presentation was in English. I think I successfully made people feel a taste of my research. I noticed that my presentation will be more attractive by incorporating my experience and what I learned from the poster presentation.

In front of GIS center at Fondren – with Kim, the head of GIS/Data Center. ~ Etsuko Ishii

What I will miss most this week after returning to Japan is the GIS center on the basement of Fondren Library. One thing that is inconvenient about my home university is the size; it is too big. When I want to ask for some advice on GIS at my home university, first I need to submit a ‘joint research application form’ to CSIS which is the similar center of Rice GIS center. After this application is accepted, then I can visit their office in Kashiwa which takes 40 minutes by train from my campus in Hongo. It’s too much hassle to follow my university’s rules; it’s a red-tape jungle. At Rice, I could easily access the GIS information center with no red-tape.

Question of the Week
What is the key factor that makes difference in atmosphere of the city? Houston and Philadelphia were completely different.

  •  A number of things. History is a key one. Houston was not established until 1839 and Philadelphia established in 1682 and was part of one of the original English colony of Pennsylvania that then became one of our first states after the Revolutionary War.  This means that there are a lot of older buildings on the East Coast, especially in the Northeast, as compared to Houston which is a much younger city.
  • What groups immigrated to that region also influence its cuisine/culture. In Pennsylvania, you’ll often find a mix of German and Italian, food. While there were many Germans that immigrated to Texas, they mainly went to the Hill Country near Austin and there was not a very large Italian immigrant population. Instead, in Houston we have Tex-Mex, lots of great Chinese and Vietnamese restaurants, and truthfully, a much more diverse food scene as we have people who’ve moved to Houston from all over the world.  So, in Houston, while Spanish, Chinese, and Vietnamese heritage may be the easiest to see there is no one dominant group/feel to the city because it has brought together people from around the world and around the U.S. who move here for our good economy, inexpensive housing, and diversity.
  • Also, the economy of the area influences the personality/culture of the city.  Philadelphia was once a manufacturing hub for the U.S.; home to many factories in the late 1800s and throughout the 19th century. It was also home to the steel industry and one of the best-known companies was actually Bethlehem Steel.  So, the region had a very working-class, manufacturing, industrial feel.  However, starting in the 1970s many of these plants started to close down as factories moved to other states in the U.S. that had cheaper wages and operating costs, or even overseas.  The steel industry in the U.S. is almost completely closed now due to global competition.  So, areas like Pennsylvania ended up having high unemployment in the 1980s and 1990s and it is only recently, in the past 20 years or so, that some cities there like Pittsburgh have really re-branded themselves and their economies to now be leaders in robotics, health care, and biomedical research.
  • Some cities/regions do this well; changing their economies and workforce to reflect the changing times.  Others struggle and there are challenges as many people in the region miss the ‘good old days’ when someone could graduate from high school and get a good-paying factory job that could support a family with just the husband working. So. there are regional inequalities in terms of economic opportunity in the U.S. that relate to the historical industries in that region and whether they have adapted their economy to the current times.  Rural regions in the U.S. often struggle with lack of good jobs and this is why people must often move to the cities to find work; similar in many ways to what has happened in rural Japan.
  • In Houston, because it was a new city and home to one of the busiest ports in the U.S. we have always had a very diverse economy that has attracted people from around the U.S. and the world.  The port means we bring in a lot of goods that are then shipped all across the U.S. via our train and freeway networks via truck.  We have the largest medical center in the world which brings doctors and nurses and patients from around the world.  The oil and gas industry and petroleum processing industry has also been a key driver of Houston’s growth. These diverse industries have led to the need for lots of service related jobs such as in education (to teach the children of the workers), finance (so people can buy homes), administrative services (to support business), restaurants (to feed people), and stores (so people can buy things), and so on.  This means our economy is hugely diverse, as compared to other parts of the U.S., and during the great recession that occurred following 2001 and also falling the housing bubble in 2008 Houston fared very well.  It was still hiring at a time when lots of companies in other parts of the U.S. were laying (firing) workers.
  • Houston is a city of newcomers that has become known for being very welcoming and the diversity is a strength; but it does mean that we don’t have such a specific personality when it comes to things like food as some other cities and regions.  But this is also why Houstonians are known for being so friendly – we have to be because we are always meeting and welcoming new people.  That is a key part of our personality that I don’t see ever changing.
  • For more on this see:

Return to Top

Final Report & Tips for Future Participants

I am almost a different person from what I was when I boarded the plane for Houston. I could not believe when Sarah told us about re-entry shock at her final lecture at Lehigh University. I thought I knew what had occurred to me, which turned out to be completely wrong. I did not notice my changes until I met my friends in the university for the first time in two months. I feel there are now many gaps with the rest of my class, especially on career formulation. The Nakatani RIES Fellowship program presented plentiful options for more to consider after I finish my bachelor’s degree. This is what I want to share with my friends the most. In Japan, and for most students in my department, they plan to graduate with their master degree and then get a job in industry. Some students who really like doing research and want to be a professor pursue PhD degrees, which makes it difficult to get a job outside academia. However, there are much more invisible options! They should expand their viewpoint as I did through this summer.

With my family, I talked mainly about the culture and life in the US. Our family tends to travel Europe or Asia Pacific regions, so my adventures in the US sounded exiting to them. What they like most is food. That’s worldwide!

For professors in my department, I reported that I had a wonderful experiences during this summer. I could not discuss about more detail, as I do not have close contact with them. For future host professors and/or employers, there are so many things to say. I will emphasize which skills I developed through Nakatani RIES Fellowship. They can be both general skills, e.g. how to make a research poster, and more specific skills, e.g. using scikit-learn package in Python for machine learning; it depends on what kind of job I will apply for. I have already trie to make my resume and cover letter more attractive by using the Nakatani RIE Program for apply to spring 2018 internships. (I’m looking for undergraduate research internship, maybe in Hong Kong)

Tips For Future Participants
Don’t hesitate to jump into new environment. Don’t be afraid of being surrounded by strange things. It may be uncomfortable for you at first, but you will enhance yourself if you could overcome these challenging situations. Moreover, you don’t have to solve difficult problems all by yourself. There are many helpful resources and kind people to support you. Health and Safety first- enjoy your summer at Rice!

Final Question of the Week
I want to know good resources to ask to correct my English writings.

  • As an alumnus, you can always email Sarah for assistance reviewing your English essays for applying to graduate schools or other programs such as spring internships. We can also review and comment on your updated resume if you’d like too.
  • You may want to find out if your home university has an English writing center or program (like the CWOVC) that could help you.  What is offered may vary by your university and faculty/academic department so you’ll need to investigate what is open to you there.
  • You may also want to review the following pages of our website which have information on communication skills including presentations, poster development, and writing:

Follow-on Project
I’m thinking to talk with Studying Abroad Office in School of Engineering and/or in Komaba, the campus for freshmen and sophomore students. There is also an office for Gender Equality which holds many events to promote my university to girls in high school, and I think I can participate some of them to invite some future STEM women researcher to Nakatani Program.

Return to Top

Print Friendly