Last year, I established Cisco’s Data Science and AI for Good initiative as a channel for Cisconians to give back pro bono, using their professional expertise to help nonprofits make the world a better place through data and analytics. Almost a year later, in collaboration with Save.org, Reportingonsuicide.org, and The Erika Legacy Foundation, we’re excited to release https://reportingonsuicide.cisco.com/.
You’re probably wondering why we chose what appears to be a niche cause. On the surface, reporting, blogging, posting on social media or talking about suicide may not seem like a life- saving or life-endangering task. Yet HOW an individual’s death by suicide is discussed and reported is a contributing factor to whether the victim’s death will likely be followed by further tragedy. So much so that the World Health Organization (WHO) established media adoption of Reporting on Suicide Guidelines as one of 7 priority areas for suicide prevention. The WHO pamphlet for media reporting on suicide is publicly available here.
These life-saving guidelines, available on reportingonsuicide.org outline how each of us can take a proactive role in suicide prevention by changing the way we speak and write about the subject. Looking at the bigger picture, when we began this effort, we realized that we could benefit all other suicide prevention efforts by helping to de-stigmatize mental illness through language. Once we understood how the language we as team members formerly used surrounding suicide contributed to the stigmatization of mental illness, it became clear that democratizing the suicide reporting guidelines wasn’t just about the media; it was an opportunity to raise awareness and understanding of this global health problem.
While none of the brilliant minds behind this effort did so for the recognition (it was a labor of love which took over many of our nights and weekends), they exemplify how each of us can make a difference- and on that note, it’s my honor to interview a key contributor to this effort, Edgar Murillo.
J: Why did you choose to donate your time and expertise to data science for suicide prevention?
E: In general, I try to use all of my Time2Give at Cisco [the 40 hours per year of paid leave Cisco gives each employee to volunteer]. When I was at the Data Symposium, I learned about Cisco’s Data Science and AI for Good program. At the time, I didn’t know what the cause would be but the opportunity just registered with me. When I got the notification that suicide prevention was one of the projects, I decided to join because I think suicide affects us all in some way. In my case, a person who’s very close to me suffers from bi-polar disorder and she has actually considered suicide several times. One time, she attempted. What concerns me is what triggers the action of suicide as opposed to ideation? With this project, I hope we can help not just the person who is suffering from mental illness but also the people who love and support them.
J: What is it like to be close to someone who suffers from bi-polar disorder and suicidal ideation?
E: Like everyone, I have my problems, phobias, and fears, but the way I deal with them is very different from the person in my life with bi-polar disorder because I’m able to manage those situations. For me, it’s very difficult to understand why someone would consider killing herself. But you have to be empathetic and that’s one of the things that I learned- that people see life very differently. The most difficult part was getting to be able to say “OK, now I understand how you think and now I see a different perspective.” Tt’s still difficult to say “it makes sense” but now I understand that it’s not something that she can control. It’s an illness, right? The illness makes them think differently and you have to understand that it’s not the person; its actually the illness that is controlling the way they think.
J: What advice would you give to someone like you who is logical and analytical as they try to understand how someone else could think in these illogical ways?
E: People ask me “how could she think of killing herself? She has everything. She has a loving family, no financial problems, etc.” When it comes to illness, we understand how people can have heart problems or diabetes and we understand how that affects them. Sometimes we forget that the brain is another organ. The brain also malfunctions. It’s actually the most complicated organ. The difference is that physically you look fine, however it impacts your behavior and your thoughts. Remember that the brain is just another organ and can also suffer from illness.
J: If I were to make a technical analogy, would you say that a person dealing with mental illness is similar to a computer with a damaged CPU trying to diagnose its own issue?
E: Yes, because one of the issue I’ve seen with bi-polar disorder is that when people are on the mania side, they think that they’re not sick- and the problem is that their mental illness is causing them to think they’re healthy. I agree- if a CPU is not working or has a bug, it is going to be difficult or impossible for it to realize or diagnose its own problem. Absolutely.
J: Continuing with this analogy, would you say that you need another device to run the diagnostic to get an accurate reading?
E: Yes, you need an external perspective. But I usually don’t try to put this in technical terms but rather in human terms. You need to have an external source running the diagnostics, giving you the feedback so the people can be cognizant of their own thoughts. And there’s a lot of professional help that can do that for you.
J: What benefits do you see of this effort?
E: I realized several things on this project; it was an education. I was one of the people who used to say ‘committed suicide’ for example. So, I think the main benefit is that we’re educating a lot of people on suicide. Not just the media, but ourselves. When I have conversations with other people, I share what I’ve learned. I think that has a multiplying effect. And of course the media has more impact. If we can help just one media outlet adopt the WHO guidelines, they can educate a lot of people.
J: You’ve given your incredible depth of technical skills to this project. Can you elaborate on that?
E: My role in Cisco is as a Solutions Architect- what we do is understand the problem that needs to be solved. Sometimes it’s a business problem. Sometimes it’s a technical problem- and then we define the elements to integrate into a solution that address the problem. I’m kind of a problem solver. That’s my role. I started my career a long time ago as a software developer, then moved into networking. Now that Cisco is moving back into software, I dusted off my old development skills and started learning Python. Combining the software development skills with the data science skills I started developing 3 years ago was very helpful in looking at the problem, suicide prevention, from a different perspective. At the beginning, I didn’t know how I would be able to help. When we kicked off the project, I saw that we had real data scientists on the team. One of my strong suits is data wrangling- getting, cleaning, and preparing data for other people to consume and then develop models with, and that’s how I had the idea to create the UI to capture the training data.
J: Can you describe the internal Cisco UI you developed for other volunteers to label the training data for the readers?
E: In machine learning, you have different ways of creating models- supervised and unsupervised. If you want to train a model, you have to feed it with training data. We needed to be able to label the training data, the articles, as following or not following the WHO guidelines by category. For example-does the article not provide guidance for where to get help? Does it use the term “commit(ted) suicide” or refer to suicide as a sin? I created an application that first would extract articles using an API. Then, a database and the UI so other volunteers could label each article and feed it back into the database. The UI made it possible for non-technical people to review and label the articles without needing to develop technical skills. The goal was to make it as easy to use as possible.
J: What do you recommend to people who want to start using their technical skills to give back like you do?
E: Register for an AI and data science for good program. And remember for suicide prevention that you’re not just helping the person going through this situation and suffering from mental illness. You’re helping all of the people around them.
Editor’s note: Jennifer is a speaker for ODSC Europe 2020 this September. Check out her talk, “Data Science for Suicide Prevention,” then!