Search

In 2016, a team of earth scientists directly engaged a team of computer scientists to identify cyberinfrastructure (CI) approaches that would speed up an earth science workflow. This paper describes the evolution of that workflow as the two teams bridged CI and an image segmentation algorithm to do large scale earth science research. The Pacific Research Platform (PRP) and The Cognitive Hardware and Software Ecosystem Community Infrastructure (CHASE-CI) resources were used to significantly decreased the earth science workflow's wall-clock time from 19.5 days to 53 minutes. The improvement in wall-clock time comes from the use of network appliances, improved image segmentation, deployment of a containerized workflow, and the increase in CI experience and training for the earth scientists. This paper presents a description of the evolving innovations used to improve the workflow, bottlenecks identified within each workflow version, and improvements made within each version of the workflow, over a three-year time period.

We present Malasakit 2.0 (meaning “sincere care” in Filipino), an inclusive, multilingual participatory online platform with feature phone integration for collecting and analyzing quantitative and qualitative textual and audio data on disaster risk reduction (DRR) strategies. Malasakit 2.0 introduces interactive voice response (IVR) to support collection of audio data via feature phone. Malasakit utilizes peer-to-peer collaborative evaluation to identify and prioritize local DRR strategies. We present results from four field tests where 261 participants provided 1,582 evaluations of current DRR strategies, and over 950 peer-to-peer evaluations on 280 textual and audio suggestions for how local government (i.e., barangays) could better support vulnerable groups (e.g., elderly, women, children, and people with disabilities) during typhoons and floods. Results suggest that individuals who engage in disaster drills are also likely to participate in their barangay's clean-up drives to reduce flooding risk by clearing drainage pathways and that those who participate in disaster drills are also likely to have enough emergency supplies for a disaster. High-rated suggestions for DRR strategies for vulnerable groups emphasize the need for communities to establish response teams that prioritize reaching out to vulnerable groups for coordination during a disaster. Malasakit can be accessed at tiny.cc/Malasakit2.

While NSF's recent Campus Cyberinfrastructure investments have catalyzed an enormous leap in campus networking capabilities, it remains necessary to integrate these capabilities into the routine workflows of researchers transferring data among their remote collaborators, data repositories, and visualization facilities. The Pacific Research Platform (PRP) is a program to develop a science-driven, data-centric "freeway system" by federating campus Science DMZs into a regional Science DMZ. Research collaborations across PRP partners serve as use cases for deploying the hardware/software and addressing security/policy/social issues to achieve high-performance data transfers between researchers. The regional PRP represents a manageable scale for initial deployment efforts and capturing lessons learned, but the PRP also informs national deployments --i.e., an eventual National Research Platform (NRP). The recent First NRP Workshop included many campuses and networking organizations and addressed scientific requirements, scaling approaches, and socio-technical issues to extend the PRP to a national level; a second NRP Workshop will be held in August 2018. In this paper, we describe efforts to build the PRP, the social and technical approaches, the domain science applications of PRP's capabilities, and results of the recent NRP workshop.

In this study, we examine suggestions for disaster risk reduction strategies provided by residents in selected disaster-prone areas in the Philippines. The study utilizes 976 suggestions on how their barangay can help them better prepare for a disaster. These were collected through Malasakit, an e-participation platform designed by University of California, Berkeley and National University (Philippines) to engage community participation in gathering qualitative and quantitative data. Analyses were conducted through biterm topic modeling (BTM) and word embedding using gensim. For better accuracy, data preprocessing was performed to remove irrelevant or noisy data. Based on the BTM result, we identified the following important codes: preparedness, disaster, awareness, community, help, seminars, kanal (canal), linisin (clean), drainage, garbage, basura (garbage). Analyses of the topic models show that disaster preparedness is an integral part in disaster risk reduction by improving solid waste management, providing seminars for public awareness and evacuation preparation. A word intrusion test was conducted where BTM scored 55.71% which implies strong cohesion of the words with their topics. For word embedding, we drilled down on the following words: community, preparedness, emergency, barangay (village), help, kanal (drainage), basura (garbage), awareness, seminars, information. The word2vec results has a cosine similarity score of 0.902 which implies strong relatedness of each word. The result shows that the participants give importance to community preparedness for emergency, helping the barangay in clean-up drive, and awareness through seminars and information dissemination.

In this paper, we explore the task of using topic modeling as an automated approach to analyze opinions, ideas, and other inputs on disaster risk reduction (DRR) gathered from local communities that have been victims of Philippine disasters. In particular, we used Latent Dirichlet Allocation (LDA) algorithm and k-means clustering with TF-IDF, and examined topics surfacing from such outputs. Our topics show that respondents from these disaster-stricken communities focus their concerns more on improving their barangay's disaster response and preparedness. The results of both LDA and k-means clustering with TF-IDF showed similarity up to a certain degree. The resulting analyses can be presented to stakeholders towards developing policies and procedures to mitigate the effects of disasters. Future works include increasing the data size and automating topic labeling.

We describe Malasakit 1.0 (meaning “sincere care” in Filipino), a customizable participatory assessment platform that collects and streamlines quantitative and qualitative analyses and insights of disaster risk reduction (DRR) strategies. While supervised classification approaches offer opportunity to understand qualitative textual suggestions, those methods break down in areas like the Philippines, home to 187 languages and nuances in varying socioeconomic contexts. Instead, Malasakit uses dimensionality reduction and peer-to-peer evaluation on qualitative textual suggestions to identify locally appropriate DRR strategies. We present results from 12 field tests conducted in eight distinct geographic locations in the Philippines. 998 participants provided 7,157 evaluations on flood and typhoon preparedness and 2,481 peer-to-peer ratings on 896 textual suggestions for how local government could improve DRR strategies. Results suggest that female participants are more confident than males in their community's ability to recover from a major typhoon. High-rated textual suggestions focus on issuing immediate early warnings and cleaning drainages to reduce flooding. Malasakit can be accessed at tiny.cc/malasakit.

M-CAFE 2.0 is an online and mobile platform that uses collaborative filtering to collect and organize student feedback each week throughout a MOOC or an on-campus course to facilitate mid-course corrections by instructors. M-CAFE 2.0 encourages students to assess course content, structure, and suggestions provided by their peers. It requires minimal extra effort from instructors and is anonymous and separate from all student records. We present results from three pilot studies from on-campus undergraduate courses with 1,211 evaluations and 5,221 peer-ratings from 169 students. Results suggest that comparative plots of past ratings, topic tags and peer-to-peer anonymous suggestion evaluations are valuable in promoting credible and diverse course evaluation. M-CAFE 2.0 is available at m-cafe.org.

The design and assessment of development initiatives is increasingly participatory, where decision makers consider feedback from affected populations. While digital data collection facilitates faster and more reliable analysis, existing data collection tools are not optimized for unstructured qualitative (textual) data and peer-to- peer participant collaboration. In this paper, we propose a system called the Development Collaborative Assessment and Feedback Engine version 1.0 (DevCAFE), a customizable participatory assessment platform that collects and integrates quantitative assessment, qualitative feedback and peer-to-peer collaborative filtering. DevCAFE incorporates a library of statistical analyses for researchers to quickly identify quantitative and qualitative trends while collecting field data. DevCAFE can run on any mobile device with a web-browser and can work with or without Internet connectivity. We present results from two pilot projects: (1) 137 participants evaluating family planning education trainings at three Nutrition Education Centers in rural Uganda, and (2) 4,518 participants evaluating policy priorities for elected leaders in the June 2015 Mexico mid-term elections. DevCAFE collected over 19,000 peer-to-peer ratings of 336 submitted ideas. Feedback gathered through DevCAFE enabled targeted reforms to the family planning efforts in Uganda and the need for increased government attention to public safety in Mexico. Case studies and interactive demos are available at: http://opinion.berkeley.edu/devcafe/

During MOOCs and large on-campus courses with limited face-to-face interaction between students and instructors, assessing and improving teaching effectiveness is challenging. In a 2014 study on course-monitoring methods for MOOCs [30], qualitative (textual) input was found to be the most useful. Two challenges in collecting such input for ongoing course evaluation are insuring student confidentiality and developing a platform that incentivizes and manages input from many students. To collect and manage ongoing ("just-in-time") student feedback while maintaining student confidentiality, we designed the MOOC Collaborative Assessment and Feedback Engine (M-CAFE 1.0). This mobile-friendly platform encourages students to check in weekly to numerically assess their own performance, provide textual ideas about how the course might be improved, and rate ideas suggested by other students. For instructors, M-CAFE 1.0 displays ongoing trends and highlights potentially valuable ideas based on collaborative filtering. We describe case studies with two EdX MOOCs and one on-campus undergraduate course. This report summarizes data and system performance on over 500 textual ideas with over 8000 ratings. Details at http://m-cafe.org.