1 -MIT Report on the 1st MIT-Tsukuba Joint-Workshop on Data Systems Science
towards Social and Business Innovations
1 -MIT

Bunkyo-School Building, Tokyo Campus, University of Tsukuba 3-29-1 Otsuka, Bunkyo-ku, Tokyo 112-0012, Japan
Faculty of Business Sciences, University of Tsukuba MIT Institute for Data, Systems, and Society (IDSS)
Graduate School of Business Sciences, University of Tsukuba
© 2020 University of Tsukuba
Organizing committee /
Faculty of Business Sciences, University of Tsukuba /
Yuji Yamada / Setsuya Kueahashi / Yasufumi Saruwatari / Naoki Makimoto /
Institute for Data, Systems, and Society, Massachusetts Institute of Technology
Ali Jadbabaie
This workshop is supported by the following grants and project: ()
Grant-in-Aid for Scientific Research (A) 16H01833 and Challenging Research (Exploratory) 19K22024 from Japan Society for the Promotion of Science (JSPS) JSPS (A) 16H01833 ()19K22024
Pre-Strategic Initiatives (Research Base-Building) program, adopted project in FY2018, University of Tsukuba 2018
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-1-

/Workshop summary /Workshop program /Opening Remarks /Abstracts & Slides 1st Session: Data management and social influences 2nd Session: Networks and markets 3rd Session: Data science and business innovations 4th Session: Data-driven decisions 5th Session: Sports analytics and management /Closing Remarks /
/About Executive Support Organizations and Project 93
/ Workshop summary

2 1 21 22 (MIT) IDSS (Institute for Data, Systems, and Society) Ali Jadbabaie , Peko Hosoi , Dean Eckles 3 3 2 1 2
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-2-
In data science research for social implementation, it is necessary to predict and use future trends from observable data and information. However, because data resulting from market transactions or business activities are based on human decisions, it is difficult to completely reproduce the true phenomenon even with a sophisticated model.
To complement such uncertainties and use them in actual business situations, we need to understand the underlying knowledge and background of the data, although their use will be limited unless we have an opportunity to share that knowledge. Moreover, multifaceted data should be utilized for business innovation, and it is considered indispensable to explore data analysis approaches including prerequisite domain knowledge. In this workshop, we refer to the research field using data domain knowledge for analyzing and modeling data (i.e., data systems science) and discuss various topics in different fields, including social networks, power systems, marketing, sports businesses, data sharing techniques, machine learning, and agent simulations.
This workshop was held successfully at the Bunkyo School Building, Tokyo Campus, University of Tsukuba, from January 21 to 22, 2020. During this period, three keynote speeches were given by Professor Ali Jadbabaie, Associate Director of the Institute for Data, Systems, and Society (IDSS), Professor Peko Hosoi, Associate Dean of the Engineering School, and Professor Dean Eckles, the Sloan Business School, Massachusetts Institute of Technology (MIT). Professors Osawa, Toriumi, and Tanaka from the Graduate School of Engineering, the University of Tokyo, also gave invited lectures. The workshop also included speakers from the University of Tsukuba: Yamada, Ban, and Kurahashi from the Faculty of Business Sciences, Sato and Sano from the Faculty of Engineering, Information, and Systems, and Takahashi from the Faculty of Health and Sport Sciences. There was a lively debate during the two- day presentation and corresponding Q&A session that allowed us to recognize important themes in business science research and to continue further interactions.
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-3-
Workshop program / Day1 Monday, January 20
10:00-10:10 Opening Remarks 1st Session: Data management and social influence / Chair: Yasufumi Saruwatari 10:10-11:10 Dean Eckles
Massachusetts Institute of Technology, USA “Seeding with limited or costly network information”
11:15-11:55 Yukie Sano
Faculty of Engineering, Information and Systems, University of Tsukuba, Japan “Information spreading on social media: Network analysis and simulation”
11:55-12:35 Masataka Ban
Faculty of Business Sciences, University of Tsukuba, Japan “A new inventory classification criterion for retail CRM”
2nd Session: Networks and markets / Chair: Setsuya Kurahashi
13:45-14:45 Ali Jadbabaie
14:50-15:30 Kenji Tanaka
Department of Technology Management for Innovation, The University of Tokyo, Japan “A Concept Proposal for Peer-to-Peer Power Exchange for Internet-of-Energy era”
15:30-16:10 Yuji Yamada
Faculty of Business Sciences, University of Tsukuba, Japan “Creating prediction error insurance market for renewable energy trading”
3rd session: Data science and business innovations / Chair: Yuji Yamada
16:30-17:10 Mika Sato-llic() )
Faculty of Engineering, Information and Systems, University of Tsukuba, Japan “Statistical Data Science Based on Soft Computing”
-4-
17:10-17:50 Yukio Ohsawa
Department of Systems Innovation, The University of Tokyo, Japan “Create an Innovators' Marketplace with Data Jackets”
Day2 Tuesday, January 21
10:00-10:15 Ali Jadbabaie
10:15-10:55 Fujio Toriumi
Department of Systems Innovation, The University of Tokyo, Japan “How to design rewarding systems for Consumer Generated Media”
11:05-11:45 Setsuya Kurahashi
Faculty of Business Sciences, University of Tsukuba, Japan “Model-based Policy Making: Health policy, Electricity market and Urban dynamics.”
5th Session: Sports analytics and management / Chair: Naoki Makimoto
13:00-14:00 Peko Hosoi
Massachusetts Institute of Technology, USA “Sports in the Age of Data”
14:05-14:45 Yoshio Takahashi
Faculty of Health and Sport Sciences, University of Tsukuba, Japan “Innovation of Sports by New Technologies and Data Systems Science in Japan”
14:45-15:00 Closing Remarks: Hideo Kigoshi ), Vice President and Executive Director for Research, University of Tsukuba, Japan
-5-
Plenary speakers /
JR East Professor of Engineering / Department of Civil and Environmental Engineering / Associate Director, Institute for Data, Systems, and Society (IDSS) / Director, Sociotechnical Systems Research Center (SSRC) / Head, Social and Engineering Systems (SES) PhD Program
KDD Career Development Professor in Communications and Technology / Associate professor of marketing in the MIT Sloan School of Management, and affiliated faculty at the MIT Institute for Data, Systems & Society (IDSS)
Associate Dean of Engineering / Neil and Jane Pappalardo Professor of Mechanical Engineering / Professor of Mechanical Engineering
-6-
List of speakers /
Masataka Ban, Faculty of Business Sciences, University of Tsukuba, Japan Dean Eckles, Massachusetts Institute of Technology, USA Peko Hosoi, Massachusetts Institute of Technology, USA Ali Jadbabaie, Massachusetts Institute of Technology, USA Setsuya Kurahashi, Faculty of Business Sciences, University of Tsukuba, Japan Yukio Ohsawa, Department of Systems Innovation, The University of Tokyo, Japan Yukie Sano, Faculty of Engineering, Information and Systems, University of Tsukuba, Japan Mika Sato-Ilic, Faculty of Engineering, Information and Systems, University of Tsukuba, Japan Yoshio Takahashi, Faculty of Health and Sport Sciences, University of Tsukuba, Japan Kenji Tanaka, Department of Technology Management for Innovation, The University of Tokyo, Japan Fujio Toriumi, Department of Systems Innovation, The University of Tokyo, Japan Yuji Yamada, Faculty of Business Sciences, University of Tsukuba, Japan
-7-
Opening Remarks /
Yuji Yamada / , Dean of Faculty of Business Sciences, University of Tsukuba
Welcome to the Myogadani School Building of University of Tsukuba, Tokyo Campus. On behalf of all the faculty members and staff of the Tokyo campus, I can say it is our pleasure to host the 1st MIT-Tsukuba Joint-Workshop on Data Systems Science towards Social and Business Innovations. My name is Yuji Yamada. I am the dean of the Faculty of Business Sciences, and in charge of the facultys general research planning and affairs. I hope you enjoy your visit here today.
Before starting the workshop, I would like to introduce the University of Tsukuba and the Faculty of Business Sciences briefly. The main campus of the University of Tsukuba is located in the Tsukuba Science City in Ibaraki Prefecture, which is about 90 min north from here by train. It was founded in 1973, due to the relocation of its antecedent, the Tokyo University of Education, to the Tsukuba area, as part of the new concept of a comprehensive university in Japan to be established under a countrywide university reform plan. The University of Tsukuba covers a wide spectrum of academic fields, from humanities to science, engineering, medicine and physical education, facilitating in-depth research and education. At the same time, it actively pursues interdisciplinary studies which do not fall into conventional academic classifications, and has featured openness with new systems for education and research under a new university administration.
The ancestor of The University of Tsukuba, the Tokyo Higher Normal School, has a long history since its establishment in 1872. It was actually located here, in this place. From its creation to now, three well-known Japanese recipients of the Nobel Prize, Dr. Sin-itiro Tomonaga, Dr. Leo Esaki and Dr. Hideki Shirakawa, have been associated with the university. The Tokyo campus of the University of Tsukuba has a different history. In 1989, after the university relocation to Tsukuba area, we started a graduate school for working people in Tokyo using the facilities of the Tokyo Education University. It was the first attempt among all national universities in Japan to offer evening graduate courses, which is the main platform for the education of working people. Since its opening, the students have not only been active
-8-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
employees in various fields, but also active students conducting research. The Faculty of Business Sciences is the organization of faculty members who are in charge of Graduate Programs here in Tokyo Campus for systems management, business law, international businesses, and legal studies. We also conduct research in the fields of economics and law to solve the business issues of the “Global Network Age” through scientific perspectives such as the data systems science approach. In this workshop, there are 3 talks from members of our group, Masataka Ban, Setsuya Kurahashi, and myself.
Let me mention that there are a total of 10 faculties at the University of Tsukuba: Humanities and Social Sciences; Business Sciences; Pure and Applied Sciences; Engineering, Information and Systems; Life and Environmental Sciences; Human Sciences; Health and Sport Sciences; Art and Design; Medicine; and Library, Information and Media Sciences. In this workshop, we have 2 talks from the Faculty of Engineering, Information and Systems by Professors Yukie Sano and Mika Sato-Ilic and one from the Faculty of Health and Sport Sciences by Yoshio Takahashi.
Also, special thanks to the invited speakers from the University of Tokyo, Professors Yukio Ohsawa, Kenji Tanaka, and Fujio Toriumi, as well as all those in attendance at this workshop. Finally, we would like to express our sincere thanks to the three plenary speakers from MIT, Professors Dean Eckles, Peko Hosoi, and Ali Jadbabaie. I hope all of you find the discussions at this workshop useful for your future research and enjoy your visit to our Tokyo campus. Thank you for your attention, and now let us start the workshop.
Opening Remarks /
Yuji Yamada / Dean of Faculty of Business Sciences
/
Abstracts & Slides /
You can download published slides in this book from this event site.

(The first paper https://arxiv.org/abs/1809.09561 is joint work with Alex Chin and Johan Ugander, and the second https://arxiv.org/abs/1905.04325 is joint work with Hossein Esfandiari, Elchanan Mossel, and M. Amin Rahimian.)
1st Session: Data management and social influence Plenary Talk 1 Dean Eckles “Seeding with limited or costly network information”
When a behavior (e.g., product adoption) may spread through a social network, it can be advantageous to consider network structure when deciding where to seed that behavior. But what if the network is not yet observed and doing so is costly? Some recent empirical work employs methods that only rely on limited network information (e.g., by exploiting the friendship paradox). We consider these and other more sophisticated stochastic seeding strategies that likewise involve sampling information about the network. In the first part, we develop nonparametric methods for empirically evaluating stochastic seeding strategies, including by reusing existing data. This draws on the policy evaluation, dynamic treatment regime, and importance sampling literatures. We show that the proposed estimators and designs can dramatically increase precision while yielding valid inference. We apply our proposed estimators to two field experiments on insurance marketing in rural China and anti-conflict interventions in New Jersey schools. In the second part, we develop novel seeding strategies that come with guarantees about the loss compared with seeding using complete knowledge of the network. These algorithms make a bounded number of queries of the network structure and provide tight approximation guarantees for arbitrary networks. We test our algorithms on empirical network data to quantify the trade-off between the cost of obtaining more refined network information, and the benefit of the added information for guiding improved seeding strategies.
-11-
Dean Eckles Sloan School of Management
Institute for Data, Systems, and Society MIT
Joint work with Alex Chin Hossein Esfandiari Elchanan Mossel
M. Amin Rahimian Johan Ugander
Seeding problems
Influence maximization
Monotone submodular influence functions
Assumes oracle access to the influence function
require explicit and complete knowledge of the network to estimate the influence function using Monte Carlo simulations
Seeding and measuring the network Kim et al. (2015) studied spreading adoption of multivitamins and a water purification method
They measure the social networks for 32 villages in rural Honduras, but contemplate avoiding this costly survey
Randomly assign villages to random, one-hop (i.e., nomination, acquaintance), or max in-degree seeding
from Kim et al. (2015)
One-hop seeding
Proposed for vaccination –
Sample k nodes at random, select a random neighbor of each as a seed
Only requires querying k edges, but doesn’t pool information across them and offers no guarantees
One-hop is importantly stochastic Village 1
One village from Kim et al. (2015). Red is random seed set, green is nomination seed set.
Random targeting
In random targeting, all eligible seed sets are equally likely, so prand i is
characterized by the uniform probabilities
Prand i (Si = s) =
, for all s ∈ Si , i = 1, ... , n, (1)
Notice that this is independent of the graph Gi , the network structure of village i .
Consider one-hop with k = 1. Nodes sized proportional to
Phop i (Si = {v}) = 1
mi
"Seeding with limited or costly network information"
-12-
Counterfactual policy evaluation: Basic example
Consider some policy that assigns units to treatment iff Xi > c. The mean outcome under this policy is
μπ = 1 N
∑ i
Yi(zπ,i)
where zπ,i = {Xi > c}. Can estimate this from a randomized experiment with treatment
probabilities pΔ(z|x) > 1 for all z ∈ {0, 1} and x ∈ X.
μHT π =
1 N
Yi {Zi = zπ,i} pΔ(Zi |Xi)
Estimand Goal: estimate the difference in expected outcomes for each of the two targeting strategies, A and B.
Finite population version:
τfp = 1 n
n∑ i=1
] , (2)
B denote expectation over Si ∼ pi A and Si ∼ pi
B, respectively.
Superpopulation version:
Notice that τsp can be written as
τsp = EΔ
[ pA(S)− pB(S)
pΔ(S) y(S) .
Design and target distributions Bernoulli design: Zi ∼ Bernoulli(ρ), where ρ ∈ (0, 1) is the treatment
assignment probability, yield a mixture:
Si ∼ ρpA i + (1 − ρ)pB
i .
i (Si = s) + (1 − ρ)PB i (Si = s). (4)
pA i , pB
i , and pΔ i are all completely known, so we know the exact
probabilities corresponding to the observed seed sets. Let si be the observed seed set for village i and let
ai = PA i (Si = si), bi = PB
i (Si = si), δi = PΔ i (Si = si)
Estimators
wA i =
wA i
Effective sample size: Designs and estimators Effective sample size across multiple data sets with k = 2 seeds per unit.
Population effective sample size n∗ eff
Dataset n τ , Bernoulli(0.5) τ , optimal τ , off-policy
Cai et al. 150 631.72 871.16 233.36 Paluck et al. 56 351.60 539.04 128.34
AddHealth 85 319.36 448.80 131.04 Banerjee et al. 75 214.32 274.08 80.24
Chami et al. 17 37.84 47.92 9.32
Estimator provides large gains. Additional gains available from optimal design. Off-policy evaluation (from random seeding only) often beats difference-in-means with Bernoulli design.
Re-analysis of Cai et al. (2015)
Experimental setup: Farmer’s insurance in 150 rural Chinese villages. Treatment: Intensive information session about insurance product to
seed set; in first round Response: Fraction of adopters in village
All villages were assigned to a uniform random strategy (prand), but our method actually enables off-policy evaluation, letting us estimate the mean response under both prand and one-hop targeting phop.
This is thus off-policy estimation
0.01
0.10
1.00
10.00
Probabilities, weights, and outcomes in Cai et al. (2015) reanalysis.
Results for Cai et al. Effect of seeding strategy on insurance purchase
estimate (one-hop − rand) -0.0436 SE (analytic) 0.0209
SE (bootstrap) 0.0257 95% CI (analytic) [-0.0846, -0.0027]
95% CI (bootstrap) [-0.0909, 0.0088] p-value (analytic) 0.0367
p-value (Fisherian) 0.0974
Hájek estimate and inference for the difference in insurance takeup rates between one-hop and random seeding for Cai et al. (2015), which provide some evidence that one-hop seeding would have reduced adoption of insurance.
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-13-
Random seeding What about not measuring the network at all?
Akbarpour, Malladi & Saberi (2019): Network information is worth “just a few seeds more”
Seeding nodes at random > seeding nodes optimally
Average-case over the random graph inputs
Limited to cascade sizes greater than
No Information Full Information
No Information Full Information
Max in-degree, degree distribution
Influence maximization, optimal seeding
Random seeding
Max in-degree, degree distribution
Problem formulation
Independent cascade
Each edge has a one time chance to pass adoptions with probability
(prune the graph, keeping each edge with probability )
Equivalent to SIR model with deterministic I to R transition
choose the best k nodes to maximize the expected number of adopters
Limited access to the graph structure through queries
Related work Influence maximization with sampled networks
Widler, Immorlica, Rice & Tambe (2018): Can find the largest blocks in a stochastic block model
Take a random sample of nodes
Use random walks to estimate their block sizes
Seed the largest blocks uniformly at random:
Guarantees limited to SBM
Our contributions
Polynomial time algorithms with tight guarantees for general graphs using a bounded number of queries
A framework to quantify the value of network information
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-14-
Two query frameworks Spread queries
Seed a random node and see the set of (eventual) adopters
Application
Edge queries
Used in sublinear algorithms
Snowball- or RDS-like surveying
Rather than optimizing the influence function on the original graph we do so on a subgraph that is properly sampled from the original graph
Running the spreading process (or something like it) turns out to be a convenient way of sampling graphs for influence maximization
Spread queries Spread queries
Each query seeds a single node and outputs the identity of the resultant adopters.
The main idea is to use the identity of adopters from independent cascades to seed nodes with optimal approximation guarantees.
Spread queries
Spread queries
Spread queries
cascades.
To prevent overlap with already chosen seeds, discard those cascades that intersect with the already chosen seeds.
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-15-
Hardness: spreading queries
Edge queries
We query a node by asking it to reveal its edges with probability p independently at random, and proceed to query it neighbors, etc.
Starting from a random set of initial seeds we obtain multiple sub- sampled copies of the original graph.
Edge queries
Edge queries Edge queries
2+3+1=6
The value of each connected component is the number of initial nodes in that component.
The total value of the marked node is the sum of the values of its connected component in each cascade.
Choose to seed the most valuable node.
Set the value of a connected component to zero if it contains a node that is chosen as the seed.
2
1
3
less than edge queries
-16-
1
3
Use the number of reachable initial nodes to score
Spread queries
seeds queries per seed =
41,536 nodes averages degree 65.59 1,362,220 edges
Trading off seed nodes for more queries
Spread queries
We can seed less nodes with more queries keeping the performance fixed.
cascade probability 6
sp re
ad s
iz e
(m ea
20 40 60
Explore the middle-ground between complete and no information about networks
Interesting heuristic seeding strategies, which can be analyzed using existing data
Use past spread data or iterative survey to acquire network information relevant to seeding
Can quantify the value of network information: Beyond all-or- nothing to noting diminishing returns to network information
Thanks!
with Alex Chin and Johan Ugander Working paper. https://arxiv.org/abs/1809.09561
Seeding with costly network information.
with Hossein Esfandiari, Elchanan Mossel, and M. Amin Rahimian Working paper. Working paper. https://arxiv.org/abs/1905.04325
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-17-
Start of Plenary Talk 1
On the lecture
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Yukie Sano / “Information spreading on social media: Network analysis and simulation”
Information spreading is a crucial issue for society and companies. For example, under emergencies, incorrect and uncertain information may coexist, causing social confusion and sometimes causing serious damage to companies. However, it is generally very difficult to accurately capture the information spreading process. This is because unlike simple spreading processes like HIV, the relationship between the sender and the receiver plays a vital role in information sharing, and makes the spreading process more complicated. In this presentation, we first present the results of a large-scale analysis of the actual data in Japan and China on both true and false information spreading. As a common feature in both countries, the network distance in false information spreading is larger than that of true information. The deviation of the degree distribution in false information spreading networks is smaller than that of the true information. When this feature is used to determine whether the information is false or true from the network topology alone without the content, it is possible to make a determination at an early stage, e.g., five hours after the first re-postings. In addition to network analysis, we also present the results of simulations using virtual scenarios on actual networks. By conducting simulations, it becomes possible to quantitatively discuss how the spreading speed and the scale change in the absence of an influential sender (influencer).
-19-
-20-
-21-
-22-
-23-
On the lecture
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Masataka Ban / “A new inventory classification criterion for retail CRM”
In this research, we propose a new criterion for goods inventory classification to formulate retailers CRM (Customer Relationship Management) strategies. In general, a retail store manager classifies their inventory according to sales volume, average unit cost, annual dollar usage, lead time and so forth. In most cases, they are short-term performance criteria for store profitability, although a retailer using forward CRM strategy needs to evaluate the profitability based on the long-term criterion such as a customer LTV (Life- Time Value). On the other hand, retailers often manage their customers using the so- called RFM metrics, in which “R” denotes purchase recency, “F” purchase frequency, and “M” average monetary amount of purchases. The retail mangers classify the customers by the RFM to identify that customers are highly/lowly profitable, and are ready to churn. Although the RFM strongly depends on data periods, Shumittlein and Peterson (1994, Marketing Science) and their subsequent studies have developed models to estimate customer LTV with a small dependence on the RFM data period. In this work, we construct a model that the customer LTV is regressed on their purchased items, assuming that a profitable customer purchases an item leading to a high performance for store profit. Then, each regression coefficient indicates how much an item does contribute to the customer LTV. Additionally, Shumittleins LTV is calculated by a PLS statistics, where the PLS stands for “Purchase rate”, “Lifetime”, and “Spending per transaction” estimated by the RFM. We investigate how much an item does work for the PLS, and conclude that the model enables us to evaluate the long-term performance of inventory being consistent to the customer LTV. Note that more detailed explanations of the model and empirical studies will be shown in the presentation.
-25-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-26-
-27-
-28-
-29-
On the lecture
Plenary Talk 2 Ali Jadbabaie “Misinformation spreading in social networks”
Start of Plenary Talk 2
Day1 Lecture
-31-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-32-
-33-
-34-
-35-
-36-
On the lecture
Day1 Front of the venue /
-38-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Kenji Tanaka / “A Concept Proposal for Peer-to-Peer Power Exchange for Internet-of-Energy era”
The last decades have seen a huge increase of distributed energy resources. Managing this growing number of intermittent resources requires both a highly interconnected grid as well as distributed storage for balancing supply and demand. As the trend for decentralized infrastructure and hardware continues, so does the need for decentralized control systems and energy management software. We argue that market mechanisms will be key to incentivizing both consumer and prosumer to compromise in electricity trading in order to alleviate peak demands and solving the demand-response problem. However, being able to conceive a secure and decentralized control and billing system adapted to such autonomous, peer-to-peer exchanges is one of the biggest challenges of this century. We propose the applicability of Blockchain as new decentralized tool to enable P2P trading within prosumers of microgrids. To test the feasibility of such a system, we are conducting two real scale microgrid projects with households, shopping centre, and plug-in EVs. The result shows the future potential for harmonizing the gap between demand and supply with this mechanism.
-39-
“A Concept Proposal for Peer-to-Peer Power Exchange for Internet-of-Energy era”
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-40-
-41-
-42-
-43-
-44-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Yuji Yamada / “Creating prediction error insurance market for renewable energy trading”
Predicting future weather conditions is important for electricity industries with renewable energy generators to quote a next-day sales contract (i.e., day-ahead sales contract). If a prediction error exists, the market-monitoring agent has to prepare another power generation resource to immediately compensate for the shortage, resulting in an additional cost for the monitoring agent. In this context, a penalty may be required depending on the size of the prediction error, which may lead to a significant loss for renewable electric power industries. Because the main source of such losses is from prediction errors of weather conditions, they can instead effectively utilize an insurance contract (or a derivative contract) based on prediction errors of weather conditions to hedge against loss caused by prediction errors of power output for renewable energy trading. The objective of this work is to introduce an insurance market based on prediction error weather derivatives, where we focus on solar power energy trades that particularly increased in Japan. To the end, we construct a cross- hedging strategy consisting of solar radiation and temperature derivatives and propose to apply nonparametric regression techniques that enable us to find optimal derivative contracts for solar prediction errors and/or optimal contract volumes of temperature derivatives. It is thought that the development and knowledge sharing of our hedging methods are effective not only for risk management of individual power industries but also for policy decision making as to efficient risk imposition on power industries.
-45-
-46-
-47-
-48-
On the lecture
Venue Day1 Just before the opening /
On the lecture
3rd Session: Data science and business innovations
Mika Sato-Ilic / “Statistical Data Science Based on Soft Computing”
Todays vast and complex societal data require adaptable methods to meet the rapidly changing needs of data structures in many areas. Statistical data analysis conventionally played a core role in dealing with any type of data and formed the theoretical foundation for a wide range of data analysis. However, conventional data analysis dependent on statistical methods is not always adequate to handle the often complex data types that make up this data. Soft computing has a key framework to tolerate imprecise and uncertain problems with a robust and computational optimal low-cost solution. This opens the door to a new paradigm based statistical multivariate analysis called soft data analysis that is capable of solving these statistical challenges. In soft data analysis, in which conventional statistical methods and machine learning or data mining methods are combined, we have developed several models that utilize the latent scale captured by soft computing techniques to explain data. While the original scale does not have the capacity to act as the scale for complex data, a scale extracted from the data itself can deal with this data. This presentation focuses on challenging issues of statistical data analysis caused by the new vast and complex data, the general problems of conventional statistical data analysis, how our soft computing based latent- scaled models are related to these issues, and how the models solve the problems with some applications.
-51-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-52-
-53-
-54-
-55-
On the lecture
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Yukio Ohsawa / “Create an Innovators Marketplace with Data Jackets”
Data Jackets are human-made metadata for each dataset, reflecting peoples subjective or potential interests. By visualizing the relevance among DJs, participants in the market of data think and talk about why and how they should combine the corresponding datasets. Even if the owners of data may hesitate to open their data to the public, they can present the DJs in the Innovators Marketplace on Data Jackets that is a platform for innovations. Here, participants communicate to find ideas to combine/use/reuse data or future ollaborators. Furthermore, explicitly or implicitly required data can be searched by the use of tools developed on DJs, which enabled, for example, analogical inventions of data analysis methods. In this talk, the speaker shows some results on the applications of DJs to the explanation of latent dynamics of , consumption market, stock market, earthquakes, and soccer games. Thus, we show a data-mediated platform to cultivate plans in business, science, and daily activities that are now being extended to the redesigning of living and working enviroments.
-57-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-58-
-59-
-60-
-61-
-62-
On the lecture
Start of Day2
-64-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-65-
-66-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Fujio Toriumi / “How to design rewarding systems for Consumer Generated Media”
After web 2.0 era, thousands of systems that provide a platform to bridge user to user are developed which called Consumer Generated Media (CGM) such as social news sites, review sites, information sharing sites and video sharing sites. Various incentive systems are implemented in CGM to encourage those content-providers. Both functions of comments and the “like” buttons are well- known systems to reward the providers. Here comes a research question: What kind of incentives encourage to post articles on CGM? In this talk, Ill explain our agent-based model to confirm the effect of incentive systems implemented to CGM.
-67-
Generated Media. The University of Tokyo
Fujio TORIUMI
• Organizer of Special Interest Group on Computational Social Science Japan
• Research Field • Computational Social Science • AI for Societies
Result
• To encourage information sharing on CGM: • Comment-Reply systems are effective • Like is effective • “Read mark” is very effective • Effect of monetary rewards are sometime limited
What kind of CGM can survive? • There exists thousands of CGM
• Born and gone • Not all of them are well used
• Do you remember Orkut?
• What is success in CGM? • Values are created by the contents • Contents are provided by users
• Cooperative users are required
How to design CGM
• Users incur costs to generates contents • Most of users avoid paying costs to provide information
• Read Wikipedia > Edit Wikipedia • Incentives to engage in “free-ride” behavior
• CGM can be modeled as Public Goods Game • Two behaviors
• Cooperate = provide information • Defection = Not provide information (only consume information)
• Defection > Cooperation
Cost to cooperate :Benefits :Number of cooperators
Cooperate or Defect • Two choices players can do
• Cooperation, Contributors • Defect, Shirkers
• Opinions of Economists : • Contributors are doomed and vanish quickly if
• In well-mixed populations • random encounters between individuals
• A cooperative society does not emerge! • In a real : people tend to contribute! • Academic challenge
• Why does cooperation evolve?
• Direct reciprocity (Fehr, 2002) • Indirect reciprocity
• Using Tags (Nowak, 2005) • Using Reputations (Ohtsuki, 2007)
2. Selective players (don’t play to all players) • Spatial structure (Helbing, 2010) • Network (Nakamaru, 2005)
3. Extra game after playing PGG
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
“How to design rewarding systems for Consumer Generated Media”
-68-
• Metanorms game • Punishment for not punish (metanorms)
• Existence of metanorms contributes to the evolution of cooperation
Axelrod, R.: An Evolutionary Approach to Norms, American Political Science Review, Vol. 80, No. 4, pp. 1095–1111(1986)
S<B
i : lost
Metanorms game
j punishes i
k punishes j
i defects Vj
Axelrod, R.: An Evolutionary Approach to Norms, American Political Science Review, Vol. 80, No. 4, pp. 1095–1111(1986)
=3 =-1
=-2 =-9
=-2 =-9
• Can we punish free-riders? Impossible • Rewards for generating contents
• ‘Like’ button, Comments, Retweet
• What kind of rewarding systems are effective?
Reward system for CGM
• Comment-reply system • Low-cost reward system
• Like button • Read marks
• Monetary reward • Points to contribution
• Research question • Which kind of rewarding systems encourage to use CGMs?
Reward system for CGM
• Comment-reply system • Low-cost reward system
• Like button • Read marks
Meta-rewards game
j rewards i k gets C’’ j gets R’’
k doesn’t reward j
k rewards j
i cooperates Lj
i gets F others get M
k : cost i : reward
i : cost j : reward
j : cost i : reward
Fujio Toriumi, Hitoshi Yamamoto, and Isamu Okada Why do people use Social Media? - Agent-based simulation and population dynamics analysis of the evolution of
cooperation in social media The 2012 IEEE/WIC/ACM International Conference on Intelligent Agent Technology(12/2012)
RR Games on CGM Facebook Q&A Site
Cooperation Update status
Acknowledge ment
• For both rewards and meta-rewards
• The game comprises three stages 1. PGG stage(Public goods game) 2. FSG stage (First-order sanctions game) 3. SSG stage (Second-order sanctions game)
F. Toriumi, et.al: Why do people use Social Media? IAT2012 (2012)
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-69-
Find cooperation & Reward
Find rewarding & meta-reward
Simulation Settings • Simulation Steps 1,000 • Number of Agents 20 • Network Perfect Graph
• Evolving strategy (adaptive process) • Genetic Algorithm
• Selection: Roulette type • Crossover: one-point type • Mutation rate: 1%
Meta-rewards game
F. Toriumi, et.al: Why do people use Social Media? IAT2012 (2012)
Robustness of cooperation
• Change in costs for rewards • = =c •
• Change in benefits • = =r • 0 r 10
Values -3.0 1.0 c r c r
Change in benefits
Effective cost-reward ratio • Cooperation becomes dominant if the benefit for
reward exceeds the cost for reward • (benefit) (cost)
• Examples in social media, • Comments on Facebook
• Low costs, low benefits • Recipes in recipe sharing sites
• High costs, high benefits
=>Point to design new social media systems is costs to generate contents and benefits
Reward system for CGM
• Comment-reply system • Low-cost reward system
• Like button • Read marks
Meta-rewards game with like!
j rewards i k gets C’’ j gets R’’
k doesn’t reward j
k rewards j
i cooperates Lj
i gets F others get M
k : cost j : reward
i : cost j : reward
j : cost i : reward
Yuki Hirahara, Fujio Toriumi, Toshiharu Sugawara Cooperation-dominant Situations in SNS-norms Game on Complex and Facebook Networks New Generation Computing, Volume 34, Issue 3, pp 273–290 (08/2016)
j doesn’t like i
j push like! for i
Lj
-70-
Results of Model with Like-ing
• This figure indicates that the optional response of clicking “Like!”This figure indicates tha buttons was effective.
• buttons was effective. Because the probability of posting articles s BBLikekek was always larger than n BBecause the prob in the basic SNS
rob NSNS-
Reward system for CGM
• Comment-reply system • Low-cost reward system
• Like button • Read marks
Monetary reward
• Some CGM has royalty points for cooperators • Get royalty points by posting articles • Ex. TripAdviser
• RQ: Is monetary rewards effect to encourage posting information?
Monetary vs Psychological reward
• No monetary rewards • R
• Monetary rewards • Loyalty points
• These CGM have similar systems • Users can post recipes • Comments to the recipes(reward system) • Replies to the comments (meta-reward system)
Extended model
• Quality of articles: • Cost of cooperation:
• Reward of PPG:
• Reward rate: •
• Each agent use choose one CGM • They have a chance to change using CGM
Change of affiliation
j rewards i k gets C’’ j gets R’’
k doesn’t reward j
k rewards j
i cooperates Lj
i gets F others get M
k : cost j : reward
i : cost j : reward
j : cost i : reward
-71-
Discussion
• Monetary rewards are limited in some cases • Quality is important factor
• Monetary rewards become incentive to post low-quality information
• If there are only peer rewards, no one gives rewards to low quality information providers
• The quality of information in CGM become lower • Users move to another CGM when they can find only
low-quality information • Cooperators who prefer psychological rewards are satisfied
even in non-monetary reward CGM
Result
• RQ : Which kind of rewarding system encourage usage of CGM?
• Meta reward system • Comment-reply system Encourage usage of
CGM • Low-cost reward system
• Like button Encourage usage of CGM • Monetary reward
• Loyalty points for the contribution • Some times fail to encourage usage of CGM
Conclusion
• It is difficult to design CGM • How can we develop CGM with information providers
• CGM can be modeled as PGG • Use agent-based simulation to realize cooperation in PGG
• Three types of rewarding system • Meta reward system • Like or read mark system
• Many CGM have • Good system to encourage cooperation
• Monetary reward system • Fail to keep quality
On the lecture
-72-
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Setsuya Kurahashi / “Model-based Policy Making: Health policy, Electricity market and Urban dynamics”
Many significant policies of our society and economy are determined by someone day after day. However, most of the plans have been discussed and decided based on past experiences and data. Many of them estimate policy effects by analysing actual phenomena and data using statistical methods. For this method called evidence-based policymaking (EBP), this lecture proposes model-based policymaking (MBP). The MBP is designed with an agent-based model and data science techniques, and it is also called as social simulation. The model-based approach enables to design realistic phenomena as a model and predict the effect on unfolding future events due to hypotheses or activities that are difficult to experiment using computer experiments.In the field of business and sociology, data analysis as an induction method and strategy planning as a deductive method are connected. In the lecture, I will introduce the analyses of health policy for infectious diseases, electricity market and urban dynamics.
-73-
-74-
= + +
+
=
= +
= +
-75-
Change to train and walk when promotion is big
Despite any location of a facility, Emergence of a sprawl structure
Despite small promotion of street attractiveness, a sprawl structure is not emerged.
Emergence of a sprawl structure Cars are dominant = actual cities
Cars are dominant
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-76-
Wikipedia
Karlsruhe, Germany
Due to the rapidly growing popularity of car transport, the tram network was gradually dismantled in Heidelberg, but since 2006, the network was extended again with the opening of a new tram line.
Three times the number of cars can be replaced by a tram. A US Transportation Research Board analysis also showed that trams carried 40% more trips than bus service.
humantransit.org
3131313333133311313131333313333313133331131311333131
-77-
In 2014-2015, WHO confirmed that a total of 28,616 cases have been reported in Guinea, Liberia and Sierra Leone, with 11,310 deaths during the outbreak.
West African outbreak Increase over time in the cases and deaths during the 2014–2015 outbreak in West Africa
During the 16 weeks, 1331 rubella cases occured in Japan. The data from Japan is under-reported to WHO.
National Institute of Infectious Disease report, 2019
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-78-
Number of rubella patient reports by age group and genders
WHO, NIID
-79-
-80-
5th Session: Sports analytics and management
Plenary Talk 3 Peko Hosoi “Sports in the Age of Data”
In light of recent advances in data collection, sports possess a number of features that make them an ideal testing ground for new analyses and algorithms. In this talk I will describe a few studies that lie at the intersection of sports and data. The first centers on fantasy sports which have experienced a surge in popularity in the past decade. One of the consequences of this recent rapid growth is increased scrutiny surrounding the legal aspects of the games, which typically hinge on the relative roles of skill and chance in the outcome of a competition. While there are many ethical and legal arguments that enter into the debate, the answer to the skill versus chance question is grounded in mathematics. In this talk I will analyze data from daily fantasy competitions and propose a new metric to quantify the relative roles of skill and chance in games and other activities. This metric is applied to FanDuel data and to simulated seasons that are generated using Monte Carlo methods; results from real and simulated data are compared to an analytic approximation which estimates the impact of skill in contests in which players participate in a large number of games. We then apply this metric to professional sports, fantasy sports, cyclocross racing, coin flipping, and mutual fund data to determine the relative placement of all of these activities on a skill-luck spectrum. The second project I will describe is a collaboration with Major League Baseball to determine the physics behind the recent increase in the rate of home runs. In this part of the talk I will enumerate different potential drivers for the observed increase and evaluate the evidence in the data (box score, ball tracking, weather, etc) in support of each theory.
-81-
Luck, Law, and Baseball: Two Sports Stories in the Age of Data
Peko Hosoi
Step 3: Watch baseball! Score points based on athlete performance
Step 1: Form a league
Step 2: Draft your team
Recent history of fantasy sports
Daily Fantasy Sports
Legal debate: Is is gambling?
(1) Bet or wager.—The term “bet or wager”—
(E) does not include—
(ix) participation in any fantasy or simulation sports game ... in which ... no fantasy or
simulation sports team is based on the current membership of an actual team ... and that
meets the following conditions:
(II) All winning outcomes re ect the relative knowledge and skill of the participants and are determined predominantly by accumulated statistical results of the
performance of individuals (athletes in the case of sports events) in multiple real-world
sporting or other events.
Types of contest: 50/50 and head-to-head Each entry contains:
• Player ID • Game ID • # of entries • win fraction • average points scored • max points scored • …
29,299,240 entries
MLB NBA NFL NHL #plrs (H2H) 89,338 107, 796 190,562 20,802
#plrs (50/50) 130,515 158,532 312,263 30,607
Max games/plr 686 383 155 271
Max entries/plr 60,200 64,287 49,657 21,209
Detecting games of chance
1. Do players have different expected payoffs when playing the game?
2. Do observable characteristics of the player correlate with the payoffs they receive?
3. Do actions that players take in the game have statistically signi cant impacts on the payoffs that are achieved?
4. Are payoffs serially correlated (implying a persistence of skill)?
Limits of luck- and skill-dominated games
rst half win fraction
second half win fraction
P
0
0.2
0.4
0.6
0.8
1
Q
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-82-
MLB
NBA
NFL
NHL
Q
P
E[R∗] = 1− [ 1 +
number of games of the ith player
Estimate the expected value of R* (and other quantities) if a large number of seasons were
played with the same player population.
1. Monte Carlo 2. Analytical estimate
?
+23%
Drag Force:
Estimate CD three ways: 1. Direct measurement (in the lab) 2. TrackMan data 3. Change in HRs per at bat
Drag coef cient
Lloyd Smith | WSU Director, Sports Science Lab
Year New Authentication
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-83-
2016 2017 2015a 2015b
The data
TrackMan 3D Doppler radar (2016-2018) • Launch angle • Launch velocity • x(t), V(t) • Spin • Spray angle • …
2,137,634 pitches 378,701 balls in play
Method 2: TrackMan Data
m dV (s)
C ha
ng e
in d
ra g
Method 3: Change in HR rate
Q: How much would the drag coef cient have to change to account for the observed change in HR rate?
Distance Home run Shited Distance
P ro
ba bi
lit y
Method 3: Change in HR rate
Change in mean distance required to account for observed change in HRs: 5 ft (!)
Direction of motion
Lift

distribution of launch angles and launch velocities+ Convert 5 ft to CD
Direct measurement, TrackMan & expected change
2016 2017 2015a 2015b
C ha
ng e
in d
ra g
San Diego press conference
-84-
Why did the drag coef cient change? Why did the drag coef cient change?
0.003 in
ball
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-85-
On the lecture
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Yoshio Takahashi / “Innovation of Sports by New Technologies and Data Systems Science in Japan”
In the 21st century, the IT industry began to drive the economy in Japan. In the sports world, Rakuten decided to enter professional baseball in 2005, and DeNA acquired the baseball team in 2012. Originally, Japanese sports have spread mainly to club activities as part of school education. Companies also owned sports clubs for employee benefits and advertising, but they did not use them for technology development. This is very different from American sports, which have developed as entertainment business, and in Japan, there was no incentive to introduce new technology into the sports world. Against this background, data analysis for winning the Olympics and world championships was the first to draw attention by the Japan Institute of Sports Sciences. The data analyst who led the national volleyball team to a bronze medal in London Olympics launched the Japan Sports Analysts Association. On the other hand, IT continued to develop in Japan, and as a result of the successful bid for the Tokyo Olympics, the number of companies that disseminated Japans technological capabilities to the world increased in the wake of the Tokyo Olympics. The Sports Agency, which was established in 2015, positions the sports industry as a growth industry and supports sports innovation, the development of new technologies, and data science as a national sports policy. In the presentation, I would like to introduce the use of New Technologies and Data Systems Science in Japanese sports.
-87-
•
•
•
• • • •
•
•
•
•
•
•
•
•
•
• • •
•
•
•
“Innovation of Sports by New Technologies and Data Systems Science in Japan”
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-88-
-89-
-90-
Closing Remarks /
Hideo Kigoshi / , Vice President and Executive Director for Research, University of Tsukuba, Japan
I would like to thank all of you for your contributions to this workshop which I believe has been successful in presenting the picture of the current research on Data Systems Science towards Social and Business Innovations. There were five sessions in the two-day workshop, and we have discussed a variety of topics on marketing, social media, networks, electricity market and its risk management, statistical data science, innovators market place, agent-based modeling, sports, and so on. Let me emphasize that we have discussed a variety of topics, because this is actually key to having a workshop focused on data systems science. Data is spread over various fields and in each field, we need specific knowledge on the data. We could analyze the data in our own field, but we expect to develop more sophisticated methodologies by utilizing the ideas used in other fields. This is the reason why have assembled in this workshop. Also, I would like to express my gratitude to all the speakers including the
three plenary speakers from MIT IDSS, professors Dean Eckles, Peko Hosoi, and Ali Jadbabaie. Special thanks to the invited speakers from the University of Tokyo, Professors Yukio Ohsawa, Kenji Tanaka, and Fujio Toriumi as well as the speakers from our group, University of Tsukuba. Let me mention that in the period of FY2018 2020, this workshop has
been supported by Pre-Strategic Initiatives, Research Base-Building program, University of Tsukuba. The adopted project is entitled “Construction of Data Science Research Station toward Market and Open Data Based Business Innovations.” I hope that in the near future, this project will attain excellent research achievements and build a research basis in collaboration with researchers here. I hope and believe that everyone has learned something very important, and
it is evident that we all have shared interests. In addition, and most importantly, I do hope that we have established contacts and made friends who may prove beneficial scientifically as well as personally. With these words, on behalf of the faculty members and research groups of the University of Tsukuba, I thank all of you for these interesting ane pleasant workshop days. Again, thank you very much.
-91-
Number of participants /
Day1 30
Day2 32
Closing Remarks / Hideo Kigoshi / Vice President and Executive Director /
Continued discussion after closing remarks /
-92-
/ About Faculty of Business Sciences
10
Enterprise Law Studies Research GroupLegal Profession Research GroupSystems Management Research GroupInternational Management Professional Research Group 4
The Structure of the Faculty of Business Sciences
At the University of Tsukuba, reforms are implemented in the systems of education and research to plan for further functional enhancements across education, research, and management as a whole. “Faculties” have been established as a new means of organization of faculty members which is unique to this university. As a general rule, all faculty members of the University of Tsukuba are affiliated with specific “faculties.” Presently, there are 10 faculties in the University of Tsukuba, including the Faculty of Business Sciences, which is an organization of faculty members that drives forward education and research, management studies, and legal studies as their subjects of study and exist to solve the problems of a business-based society in the “global network age” from a scientific viewpoint. It encompasses four research groups:
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
/ Executive Support Organizations and Projectr
-93-
the Graduate School of Business Sciences
The Roles and Missions of the Faculty of Business Sciences
The Faculty of Business Sciences undertakes, “The social implementation and the construction of systems and standards of advanced technology in current society, where computerization, internationalization, and fluidization are progressing,” as its role. Its mission is driving forward research activities that are related to social infrastructure and market designs that are safe and sustainable and giving feedback toward real society. Through collaboration with the educational organization, i.e., the Graduate School of Business Sciences, it conducts research for solving pertinent issues in real society.
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-94-
1989 Management studiesLaw studies 3 Role 1 Role 2 Role 3
The Roles and Missions of the Graduate School of Business Sciences
The Graduate School of Business Sciences, where the faculty members affiliated with the Faculty of Business Sciences, mainly conduct educational activities. It is a graduate school for working adults that was established in 1989, and the majority of its faculty members are in charge as full-time faculty members. Lectures held in the evenings and conduct the training and re-education of highly advanced professionals related to “management studies” and “law studies,” which are both inseparable in supporting the core of any business. The Graduate School of Business Sciences bears the following three roles: “The nurturing of business professionals who possess theoretical thinking abilities and academic knowledge”; “The training of researchers that possess knowledge and judgment abilities related to business”; “The pioneering of recurrent education in which it is possible to attain professional degrees or degrees greater than Masters degrees without quitting ones job.”
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-95-
(https://idss.mit.edu/about-us/)
The mission of IDSS is to advance education and research in state-of-the-art analytical methods in information and decision systems, statistics and data science, and the social sciences, and to apply these methods to address complex societal challenges in a diverse set of areas such as finance, energy systems, urbanization, social networks, and health.
Technology advances in areas such as smart sensors, big data, communications, computing, and social networking are rapidly scaling the size and complexity of interconnected systems and networks, and at the same time are generating masses of data that can lead to new insights and understanding. Research at IDSS aims to understand and analyze data from across these systems, which present unique and substantial challenges due to scale, complexity, and the difficulties of extracting clear, actionable insights.
Our ability to understand data and develop models across complex, interconnected systems is at the core of our ability to uncover new insights and solutions. IDSS offers academic programs, some via the Statistics and Data Science Center (SDSC):
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-96-
IDSS research entities include: Laboratory for Information and Decision Systems (LIDS) Sociotechnical Systems Research Center (SSRC)
Spanning all five schools at MIT, IDSS embraces the collision and synthesis of ideas and methods from analytical disciplines including statistics, data science, information theory and inference, systems and control theory, optimization, economics, human and social behavior, and network science. These disciplines are relevant both for understanding complex systems, and for presenting design principles and architectures that allow for the systems quantification and management. IDSS seeks to integrate these areasfostering new collaborations, introducing new paradigms and abstractions, and utilizing the power of data to address societal challenges.
IDSS greatly values diversity in and inclusion of our students, faculty, and staff. The range of cultures, backgrounds, perspectives, and experiences of the individuals within IDSS all contribute to our ability to live out our mission of tackling major societal problems using innovative, holistic, data-driven approaches. Likewise, as a unit focused on improving society, we care deeply about the overall well-being of our studentsincluding their mental and physical health.
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
IDSS offers academic programs, some via the Statistics and Data Science CenterSDSC):
Doctoral Program in Social and Engineering Systems (SES) Technology and Policy Program (TPP) Undergraduate Minor in Statistics and Data Science (via SDSC) Interdisciplinary Doctoral Program in Statistics (via SDSC) IDSS has also developed two online education programs: MicroMasters in Statistics and Data Science
-97-
30
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
-98-
The Strategic Initiative is a research base at the University of Tsukuba that has a track record worthy of being called the world's highest level and has interdisciplinary abilities that make use of the characteristics of the university. To develop into a state-of-the-art research facility that opens up new academic research fields, the university adopts two or three Pre-Strategic Initiatives (Research Base-Building) every year, which have the potential to form strategic initiatives in the future. Under such support, the Faculty of Business Sciences has executed a research project titled “Construction of Data Science Research Station toward Market and Open Data Based Business Innovations.” The project was adopted as the 2018 pre-strategic initiative (research base-building), and since then, we have promoted research activities and interactions and held international workshops.
Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Bachground Objective
In recent years, the framework for market transactions and the mutual use of open data has been rapidly evolving in real business fields. For example, retail electricity liberalization has led to an era in which wholesale electricity is priced according to market principles, and the government is actively releasing open data platforms for utilizing the data in business activities. To analyze such accumulated data and connect it to new strategies, it is necessary to understand both advanced business practices and academic research at the same time. Based on this research center, the Faculty of Business Sciences aims to promote data science research to support business innovations by deepening mutual understanding between business and research in cooperation with the Graduate School of Business Sciences, University of Tsukuba.
-99-



Faculty of Business Sciences, University of Tsukuba, Tokyo, Japan
Tokyo Campus, Bunkyo-School Building