table of proposed coding projects

Mentors, please edit this wiki page, and add your ideas to the table below.

Students, please look for a project that interests you in the table below. Before emailing project mentors, please do at least one project Test and post a link to your solution on the proposal’s wiki page.

Student	Proposal	Results	Mentors	Non-R languages?
Evandro Taquary	GPU accelerated comparative phylogenetic analysis with uncertainty		Erin Hodgess, Thiago Rangel	CUDA
cdries	Improved functionality for higher order comoment estimation in PerformanceAnalytics		B Peterson, K Boudt	C++/Rcpp
lwei	Integrated Oversampling for Sequence Classification (IOSC)		Matthew Dixon , Diego Klabjan	C++
Leopoldo Catania	MSGARCH		D Ardia, K Boudt, B Peterson	C++
Samuel Borms	Sentometrics	GitHub	K Boudt, D Ardia, K Bluteau	C++
Faizan Khan	Animated interactive plots (animint)	blog, commits	TD Hocking, C Sievert	JavaScript
Rover Van	Speed optimizations for iregnet	blog, commits	TD Hocking, A Khare	C++
Devinder Kaur	Rperform: Performance analysis and visualization for R		Akash Tandon, Joshua Ulrich	shell scripting, possibly Javascript
Binxiang Ni	Sparse matrix automatic conversion in RcppArmadillo	blog, repo	Qiang Kou and Dirk Eddelbuettel	C++
Lindsay Rutter	bigPint: Big multivariate data plotted interactively		Dianne Cook, Roxane Legaie	Shiny, JavaScript, Plotly
Shubham Chaturvedi	Constrained Hierarchical Agglomerative Clustering	Github Gist, repo	N Villa-Vialaneix, P Neuvial	C++
Jialin Ma	Interactive Genome Browser in R	blog repo	Toby Dylan Hocking, Miguel Pignatelli	Javascript
Earo Wang	sugrrants: Visual methods for big temporal data		Dianne Cook, Rob Hyndman	R, plotly, ts, hts
mb706	Operator Based Machine Learning Pipeline Construction		Bernd Bischl, Lars Kotthoff
Thiloshon Nagarajah	Integrating biodiversity data curation functionality	Repo , Blog	Tomer Gueta, Yohay Carmel, Vijay Barve
Ashwin Agrawal	Biodiversity data cleaning	Github Repo	Vijay Barve, Yohay Carmel, Tomer Gueta
Coin Lewis-Beck	nimbleEcology package: ecological statistics driven by NIMBLE		Perry de Valpine, Daniel Turek
Xia Zhang	Graphical Models for Mixed Multi Modal Data	Home Page	Genevera Allen, Zhandong Liu, Michael Weylandt	C++
Alexandre Almeida	Distributional Assessments with Q-Q plots		Heike Hofmann, Adam Loy	R, plotly, shiny
Natalia da Silva	metawRite: Meta analysis update package, LSR (Living systematic review).		Heike Hofmann, Annette O’Connor	R, plotly, shiny
Robin Kohze	FireData: Connecting R to Google Firebase	Github Repo	S. Schmidt & L. Pedraza	-
Ben Ubah	control: Control Systems toolbox	REPO, Blog Post	Marijan Kostrun Hans W Borcher
Wazeer Zulfikar	Native R API for TensorFlow	Repo, Blog	Krishna Sankar, Tomasz Melcer	C++, python
Balázs Dukai	Interactive trajectory tool in rpostgisLT		Mathieu Basille, David Bucklin, Clement Calenge	SQL, Shiny, possibly python
Xin Chen	Risk and Performance Standard Errors for Serially Correlated Returns		R Douglas Martin, Aleksandr Aravkin, Daniel Hanson, Brian Peterson	C++/Rcpp
Vandit Jain	The markovchain package	gist, commits	SB Yalamanchi (main), G A. Spedicato (backup)	C++/Rcpp
Leah Price	Extending the RcppSMC Package		Dirk Eddelbuettel and Adam Johansen	C++/Rcpp
Pushpak Sarkar	Portfolio Construction and Risk Management with Unequal Returns Histories		Yindeng Jiang, R Douglas Martin, Peter Carl
Qingyue Xu	Parser for Biodiversity checklists	Github Repo	Vijay Barve, Rohit George, Thomas V., Narayani Barve
Lorenz Walthert	Noninvasive source code formatting	Home Page	Kirill Müller, Yihui Xie
Matthew Piekenbrock	Estimating the empirical Cluster Tree		Michael Hahsler, Mikhail Belkin	C++/Rcpp
Chindhanai Uthaisaad	advancing factorAnalytics		Eric Zivot, R Douglas Martin, Kjell Konis, Thomas Philips, Brian Peterson	C++/Rcpp
Luis Damiano	Bayesian Hierarchical Models in Finance	Blog Repo	B Peterson , Michael Weylandt	STAN
Jason Ge	Efficient Second order Optimization Solvers for Sparse Learning in R		Tou Zhao and Xinggou Li	C++
Zhehui Chen	Stochastic variational inference		Tou Zhao and Xinggou Li	C++
PROJECTS BELOW	NOT CHOSEN – TODO COPY TO NEXT YEARS WIKI
	MENTORS: Copy this template!
	SPArse Modeling Sofware			C++
	Multi-task learning			C++
	Caffe deep neural networks			C++
	Space time data over complex domains		Laura Sangalli, Luca Formaggia, Eardi Lila	R, C++
	Multi-Armed-Bandit		Wei Qian, Nathaniel Helwig	R, C++
	Management and comparison of MCMC runs from multiple packages		Perry de Valpine, Daniel Turek	html, possibly shiny
	Clustering Model for ordinal data		Julien Jacques & Christophe Biernacki	shiny, RcppArmadillo
	Constrained changepoint detection		TD Hocking, G Rigaill	C++
	Meucci Project		E Biceroglu, B Peterson, D Ardia	C++, potentially jupyter
	Symbolic Computation in R		Isuru Fernando	C++

Project ideas have a ‘Status’ column which describes the current status of mentor and student interest. Project ideas where no student has yet contacted mentors should be listed as ‘needs students’. Project ideas where one or more potential students are communicating with mentors should have a status of ‘potential student’ or ‘two[three,etc] potential students. You can still communicate your interest to mentors to apply to projects with status “potential student” – that implies that there is another student who has already shown some capability for that project (see below for more details on how we evaluate applications). Projects that need to identify another mentor (e.g. to find a mentor with a specific skill, or from a different institution) should be marked with a status of ‘need mentor’ and the idea page should provide details in the ‘Mentors’ section.

All student applications will be discussed by the R mentor community, and proposals will be ranked considering factors such as quality, difficulty, and impact for the R community. Slots are a finite resource granted to R by Google, and only the best proposals will get chosen. In prior years, R has received 4-5 times more applications than slots, so application quality is key.

Students, if you have an idea for an R package coding project that is not listed above, please try to find mentors by posting a description of your project idea on the r-gsoc google group. If you find mentors, feel free to add your project idea to this wiki. You should NOT submit any project applications to Google without finding 2 mentors for your project proposal.