Tag Archives: ai

Big Data to Data Science: Moving from ‘What’ to ‘How’ in MERL

Guest post by Grace Higdon

Big data is a big topic in other sectors but its application within monitoring and evaluation (M&E) is limited, with most reports focusing more on its potential rather than actual use. Our paper,  “Big Data to Data Science: Moving from ‘What’ to ‘How’ in the MERL Tech Space”  probes trends in the use of big data between 2014 and 2019 by a community of early adopters working in monitoring, evaluation, research, and learning (MERL) in the development and humanitarian sectors. We focus on how MERL practitioners actually use big data and what encourages or deters adoption.

First, we collated administrative and publicly available MERL Tech conference data from the 281 sessions accepted for presentation between 2015 and 2019. Of these, we identified 54 sessions that mentioned big data and compared trends between sessions that did and did not mention this topic. In any given year from 2015 to 2019, 16 percent to 26 percent of sessions at MERL Tech conferences were related to the topic of big data. (Conferences were held in Washington DC, London, and Johannesburg).

Our quantitative analysis was complemented by 11 qualitative key informant interviews. We selected interviewees representing diverse viewpoints (implementers, donors, MERL specialists) and a range of subject matter expertise and backgrounds. During interviews, we explored why an interviewee chose to use big data, the benefits and challenges of using big data, reflections on the use of big data in the wider MERL tech community, and opportunities for the future.

Findings

Our findings indicate that MERL practitioners are in a fragmented, experimental phase, with use and application of big data varying widely, accompanied by shifting terminologies. One interviewee noted that “big data is sort of an outmoded buzzword” with practitioners now using terms such as ‘artificial intelligence’ and ‘machine learning.’ Our analysis attempted to expand the umbrella of terminologies under which big data and related technologies might fall. Key informant interviews and conference session analysis identified four main types of technologies used to collect big data: satellites, remote sensors, mobile technology, and M&E platforms, as well as a number of other tools and methods. Additionally, our analysis surfaced six main types of tools used to analyze big data: artificial intelligence and machine learning, geospatial analysis, data mining, data visualization, data analysis software packages, and social network analysis.

Barriers to adoption

We also took an in-depth look at barriers to and enablers of use of big data within MERL, as well as benefits and drawbacks. Our analysis found that perceived benefits of big data included enhanced analytical possibilities, increased efficiency, scale, data quality, accuracy, and cost-effectiveness. Big data is contributing to improved targeting and better value for money. It is also enabling remote monitoring in areas that are difficult to access for reasons such as distance, poor infrastructure, or conflict.

Concerns about bias, privacy, and the potential for big data to magnify existing inequalities arose frequently. MERL practitioners cited a number of drawbacks and limitations that make them cautious about using big data. These include lack of trust in the data (including mistrust from members of local communities); misalignment of objectives, capacity, and resources when partnering with big data firms and the corporate sector; and ethical concerns related to privacy, bias, and magnification of inequalities. Barriers to adoption include insufficient resources, absence of relevant use cases, lack of skills for big data, difficulty in determining return on investment, and challenges in pinpointing the tangible value of using big data in MERL.

Our paper includes a series of short case studies of big data applications in MERL. Our research surfaced a need for more systematic and broader sharing of big data use cases and case studies in the development sector.

The field of Big Data is rapidly evolving, thus we expect that shifts have happened already in the field since the beginning of our research in 2018. We recommend several steps for advancing with Big Data / Data Science in the MERL Space, including:

  1. Consider. MERL Tech practitioners should examine relevant learning questions before deciding whether big data is the best tool for the MERL job at hand or whether another source or method could answer them just as well.
  2. Pilot testing of various big data approaches is needed in order to assess their utility and the value they add. Pilot testing should be collaborative; for example, an organization with strong roots at the field level might work with an agency that has technical expertise in relevant areas.
  3. Documenting. The current body of documentation is insufficient to highlight relevant use cases and identify frameworks for determining return on investment in big data for MERL work. The community should do more to document efforts, experiences, successes, and failures in academic and gray literature.
  4. Sharing. There is a hum of activity around big data in the vibrant MERL Tech community. We encourage the MERL Tech community to engage in fora such as communities of practice, salons, events, and other convenings, and to seek less typical avenues for sharing information and learning and to avoid knowledge silos.
  5. Learning. The MERL Tech space is not static; indeed, the terminology and applications of big data have shifted rapidly in the past 5 years and will continue to change over time. The MERL Tech community should participate in new training related to big data, continuing to apply critical thinking to new applications.
  6. Guiding. Big data practitioners are crossing exciting frontiers as they apply new methods to research and learning questions. These new opportunities bring significant responsibility. MERL Tech programs serve people who are often vulnerable — but whose rights and dignity deserve respect. As we move forward with using big data, we must carefully consider, implement, and share guidance for responsible use of these new applications, always honoring the people at the heart of our interventions.

Download the full paper here.

Read the other papers in the State of the Field of MERL Tech series.

3 Lessons Learned using Machine Learning to Measure Media Quality

by Samhir Vasdev, Technical Adviser for Digital Development at IREX’s Center for Applied Learning and Impact. The post 3 Lessons Learned using Machine Learning to Measure Media Quality appeared first on ICTworks.

Moving from hype to practice is an important but challenging step for ICT4D practitioners. As the technical adviser for digital development at IREX, a global development and education organization, I’ve been watching with cautious optimism as international development stakeholders begin to explore how artificial intelligence tools like machine learning can help them address problems and introduce efficiencies to amplify their impact.

So while USAID was developing their guide to making machine learning work for international development and TechChange rolled out their new course on Artificial Intelligence for International Development, we spent a few months this summer exploring whether we could put machine learning to work to measure media quality.

Of course, we didn’t turn to machine learning just for the sake of contributing to the “breathless commentary of ML proponents” (as USAID aptly puts it).

As we shared in a session with our artificial intelligence partner Lore at MERLTech DC 2018, some of our programs face a very real set of problems that could be alleviated through smarter use of digital tools.

Our Machine Learning Experiment

In our USAID-funded Media Strengthening Program in Mozambique, for example, a small team of human evaluators manually score thousands of news articles based on 18 measures of media quality.

This process is time consuming (some evaluators spend up to four hours a day reading and evaluating articles), inefficient (when staff turns over, we need to reinvest resources to train up new hires), and inconsistent (even well-trained evaluators might score articles differently).

To test whether we can make the process of measuring media quality less resource-intensive, we spent a few months training software to automatically detect one of these 18 measures of media quality: whether journalists keep their own opinions out of their news articles. The results of this experiment are very compelling:

  • The software had 95% accuracy in recognizing sentences containing opinions within the dataset of 1,200 articles.
  • The software’s ability to “learn” was evident. Anecdotally, the evaluation team noticed a marked improvement in the accuracy of the software’s suggestions after showing it only twenty sentences that had opinions. The accuracy, precision, and recall results highlighted above were achieved after only sixteen rounds of training the software.
  • Accuracy and precision increased the more that the model was trained. There is a clear relationship between the number of times the evaluators trained the software and the accuracy and precision of the results. The recall results did not improve over time as consistently.

These results, although promising, simplify some numbers and calculations. Check out our full report for details.

What does this all mean? Let’s start with the good news. The results suggest that some parts of media quality—specifically, whether an article is impartial or whether it echoes its author’s opinions—can be automatically measured by machine learning.

The software also introduces the possibility of unprecedented scale, scanning thousands of articles in seconds for this specific indicator. These implications introduce ways for media support programs to spend their limited resources more efficiently.

3 Lessons Learned from using Machine Learning

Of course, the machine learning experience was not without problems. With any cutting-edge technology, there will be lessons we can learn and share to improve everyone’s experience. Here are our three lessons learned working with machine learning:

1. Forget about being tech-literate; we need to be more problem-literate.

Defining a coherent, specific, actionable problem statement was one of the important steps of this experiment. This wasn’t easy. Hard trade-offs had to be made (Which of 18 indicators should we focus on?), and we had to focus on things we could measure in order to demonstrate efficiency games using this new approach (How much time do evaluators currently spend scoring articles?).

When planning your own machine learning project, devote plenty of time at the outset—together with your technology partner—to define the specific problem you’ll try to address. These conversations result in a deeper shared understanding of both the sector and the technology that will make the experiment more successful.

2. Take the time to communicate results effectively.

Since completing the experiment, people have asked me to explain how “accurate” the software is. But in practice, machine learning software uses different methods to define “accuracy”, which in turn can vary according to the specific model (the software we used deploys several models).

What starts off as a simple question (How accurate is our software?) can easily turn into a discussion of related concepts like precision, recall, false positives, and false negatives. We found that producing clean visuals (like this or this) became the most effective way to explain our results.

3. Start small and manage expectations.

Stakeholders with even a passing awareness of machine learning will be aware of its hype. Even now, some colleagues ask me how we “automated the entire media quality assessment process”—even though we only used machine learning to identify one of 18 indicators of media quality. To help mitigate inflated expectations, we invested a small amount into this “minimum viable product” (MVP) to prove the fundamental concept before expanding on it later.

Approaching your first machine learning project this way might help to keep expectations in line with reality, minimize risks associated with experimentation, and provide air cover for you to adjust your scope as you discover limitations or adjacent opportunities during the process.