Mapping Ideas with Open Source Handwriting-to-Text Software

hey there world

Ideas are curious things. You’re going along and suddenly… Boom! An idea. But it wasn’t really out of the blue, was it? Sure, something immediate may have sparked it. The spark came from somewhere. Where is that? How.. What is an idea? Configurations of neurons grow and represent realization..such vague explanation is saying Earth is a space rock. It is this. It’s immensely more.

Through language and emotion we navigate the world. We choose how to spend time, with whom to spend it. We choose which ideas are most interesting and seek out more. We read, browse, share. Videos, politics, religion, business, creativity.. our modern world is full of inspiration for those who explore. The idea I’ve been most interested in exploring is categorical: pondering how it is that ideas come about in the first place. Along this intellectual journey, I encounter a great breath of people, perspectives, and experiences. Playfully I explore the world of minds in order to create a perspective that is uniquely this self, this Amy.

In an effort to explore a bit of how I became myself, I’ve developed a side project over the past few years: mapping ideas. This post and accompanying a lecture at Stanford summarize the why.

Let this focus on the next step: sharing.

I write frequently. Between 2008 and June, 2016, I have filled up over 40 Moleskine Notebooks. Words, musings, aphorisms, speech outlines, sketches.. these notebooks are not mere writing, if there is such a thing. They are the evolution of the ideas I hold dear. The ideas I think have played the lead roles in my becoming who I am today.

Analyzing language for things like sentiment and keywords has for many years been possible thanks to platforms like AlchemyAPI from IBM. There is still a missing puzzle piece: a tool that automatically transcribes images of notebook pages into text files. Several tools exist but none so far have worked for me. I realize that Evernote makes writing searchable. It does not allow you to export text files.

Anyone out there interested in computer vision tools that turn writing to .txt? I would love to team up with you. Not just to make the next step in mapping ideas but to create an open source tool that anyone who writes can use to transform handwriting into a digital database.

I’m @amyleesterling on GitHub and Twitter. I’m still of course ever-searching for deeper understanding of who I am and perpetually welcome conversations with those of you on similar journeys. Ping me if this is up your curiosity alley. These are my personal notes and currently I’m brave enough to share them with people who ask but not publicly on the web just yet 🙂

A few sample images of notes below illustrate content diversity and wackiness.

Continue Reading

Maps of Ideas

mapping ideas quid neuroscience arxiv snapshot, mapping ideas, neuroscience, arxiv, quid, amy robinson, map ideas, graph ideas, mapping ideas

A universe.

We live in one. Your mind is one.

Human beings, besides generating things like science and technology and businesses, generate ideas. Thoughts. Sometimes thoughts lead to action. Life happens or rather is made by he who lives it.

I’ve been pondering: how could we explore being human from a perspective of ideas over time? Say, my personal ideas over time. A dynamic network, the questions, concepts and values that fuel who I am. Creativities and habits; discoveries and experiences.

If you were to document things you think are important or things you are curious about or wonder, what might you have after a month? A year? Eight years?

That’s how long I’ve been doing this. I’ve amassed 30 Moleskine notebooks, 3.5G or 756 voice memos, 6,000 Tweets and gigs of autotune on t-payne (don’t judge).

How could you map ideas? Here was my first attempt from 2012, delivered for Quantified Self at Stanford.

I spent hundreds of hours figuring out how to map 6 months of ideas in the form of emails to self. A collaboration built through an amazing community of Gephi devs.

Others are making strides in this arena. Watch the below TEDTalk about mapping ideas from the top 25% of TEDxTalks. From transcripts to a network you can interact with and explore. Beautiful. And insightful. How could these ideas apply to the ideas of an individual over time?

How do the things I’m interested in evolve? What new things have I learned and how have they made their way into the projects I create or things I do or learn in the future? When do ideas change how I think? After I learn something that changes how I think, it can be difficult if not impossible to retain how I thought before I realized it. Particularly over years. These deep ideas fascinate me.

So I’m exploring them. Publicly. And I’m going to make all of my personal data public someday. Still working up the nerve to put up my browsing history. The short term will see transcripts of voice memos and handwriting. We may need to create new language processing algorithms for stream of consciousness.

Publicly. If you think this is interesting, contact me and think about it with me. I’m using Quid and learning principles of graph theory, community detection, python, JSON, dealing with audio transcription, and most interestingly figuring out how to build a network/networks out of ideas.

Challenging. Exciting. Neural avalanche inducing.

Continue Reading

Quantified Curiosity 2.0

ideas, Gephi, "network visualization" "Quantified Curiosity" Amy RobinsonBack in September I gave a talk at Stanford for Quantified Self titled Quantified Curiosity (summarized in this post, which includes slides and a links to videos refrenced in the talk). Below, check out the video complete with a text transcript.

Before you watch it, think about this. Who are you? Seriously, how do you answer that question? Does who you are change over time? How? Why? When? What if you could explore these questions empirically, with data that correlates with significant events in your life, data that collectively integrates to tell a story of who you are? This is what I begin to explore with Quantified Curiosity, a network exploration into the ideas that fuel me. As of March 2013, I’ve connected with a couple academic and corporate network powerhouses to this concept a few orders of magnitude higher and deeper. More on that soon.

Over the coming months, stay tuned for the evolution of questions, new visualizations, and curiosity progress reports. A goal of this side project is to create a platform that allows anyone to explore and graph his or her ideas over time. Here’s to tackling fundamental questions! Ping me if you are interested in brainstorming. Now, on with the evolution of ideas!

Transcript with slide selections:

Quantified Curiosity brainbow Amy RobinsonI am obsessed with thinking about thinking.

My name is Amy Robinson and I am here to share Quantified Curiosity.

I am very curious how the ideas that I encounter and the new things that I discover integrate and infuse to form who I am and who I will become.

A stranger at a TED Conference once walked up to me and said “Hi Amy, What inspires you?” Besides actually making me think about what inspires me, it made me think about how the things that inspire me change over time. I am not a constant, I am very dynamic; however, it’s hard to remember how I change and to keep it in perspective.

Those 5 seconds consequently have mattered much more than just 5 seconds and I wonder if the same is true for ideas. So I’ve been tracking them.

How? I email myself “interestingness.” So when I look at say an article or write notes or watch a cool video; anything that makes me think “hm, that’s interesting,” I email it to myself. For this talk I’ve compiled 6 months of this data into..a pretty big spreadsheet and some beautiful network visualizations.

Each line is an idea, an entry, and the data has attributes like a date, a link, an ngram (which is the subject and body text of the email), it’s tagged with topics and it’s also given an interestingness ranking of 1 being low and 5 being high.

ideas, Gephi, "Quantified Curiosity" Amy RobinsonSix months worth of data came to 770 unique entries – or ideas – in 772 different topics. Once this data was organized into a spreadsheet I was able to analyze it and look at it in a completely new way.

This is a weighted graph  [below] of the most important topics of all topics that were used at least 40 times and weighted either 4, the green bar for “important,” or 5, the blue bar for “most important,” they show up on this graph. You can see based on the importance that the most prevalent topics vary. For example, the green bar most important is “journal,” which is peer reviewed literature, not my personal notes, followed by biology and neuro. Whereas if you look at the blue bar “notes,” my personal notes, come up first.

"Quantified Curiosity" Amy Robinson

photosofnotes, photos, notes, tumblr, amy robinson, quantified curiosity,You can also look at most important entries over time [graph below]. The most important entries  tend to occur in clusters. I wonder do these clusters actually correspond to something? There’s a huge cluster in February, 14 items in 3 days. They actually correspond to my starting a new side project, photos of notes, it’s a tumblr blog where I just publish photos of my notes. In that case, yes, that cluster was something real. And I wondered, is this true for the other clusters?

quantified curiosity, QS, quantified self,

Turns out, yes. In March there’s another one where 21 items occur in a period of 21 days. It corresponds to something kind of goofy that I do — lifebonus emails. I send these out now quite periodically to my friends saying, ya know, share something beautiful, inspiring, intelligent or entertaining that you’ve discovered in the past week and they get a hypothetical lifebonus. It’s goofy, it’s fun, it rocks the inbox but again the data actually corresponds to my doing something new.

How else can we actually explore this?

We were able to formulate these ideas into Gephi, a free network graphing program. The way this works: the circles are called nodes and they correspond to topics that are tagged with ideas. The size of the nodes indicate how many times they were used in tandem with other nodes. The edges – the lines between them – are the actual ideas that are co-tagged with the two different topics.

ideas, graph, Gephi, "network visualization" "Quantified Curiosity" Amy Robinson, "Quantified Self", nodes, edges

ideas, graph, Gephi, "network visualization" "Quantified Curiosity" Amy Robinson, "Quantified Self", nodes, edgesYou can run statistics in Gephi to modularize communities so based on how connected groups of nodes are relative to the overall connectivity of the whole graph and see distinct communities. For example, the blue down at the bottom is science and science-related tags. The purple is work slash health — I work[ed] in health; you can probably actually infer that by looking at the graph. The red section is TED and TED-related tags, including TEDx and video. And then the green section is “self” and there were come cool things in there like playful, curious, ideas and Quora that popped up really close to me. But this is messy. It’s hard to see 10,500 edges so what you can do is you can actually isolate individual topics.

ideas, graph, Gephi, "network visualization" "Quantified Curiosity" Amy Robinson, "Quantified Self"

The yellow dot here is the tag “ideas” within all my ideas data. You can see the little green dot sort of off to the side. It exhibits what’s called a high “betweenness centrality.” In social network graphs that represent people, those nodes that have a high betweenness centrality are the ones that bridge gaps between distinct communities. They’re interdisciplinary in a way and it made me wonder, could the same be true for ideas? Those “in between” ideas, and how can I decipher this information?

ideas, Gephi, "network visualization" "Quantified Curiosity" Amy Robinson, "Quantified Self", beautifulWe can look at the graph of “beautiful” for an example. You see there’s a purple dot right in the middle. That’s “tech” and when I actually looked at these tags, there’s a series of beautiful, scientific, technological videos, that I’ve actually compiled on my blog [here!] if you’re curious to see them. You can also zoom in on this red section that were closely tagged with “beautiful” — so “TED”, “TEDx”, “side project”, I guess it’s a good sign that the things I do for free in my spare time incite a sense of awe and beauty. “Video” was the largest in that cluster.

ideas, Gephi, "network visualization" "Quantified Curiosity" Amy Robinson, "Quantified Self", video

When I actually look at the graph of “video,” it made me wonder how we could take this information and make it interactive. Imagine you were panning through this on a computer and rather than just looking at nodes, you could actually look at the content relative to where they’re tagged and other things

Here is the tag for “self.” A lot of this was intuitive — “TED,” “science,” — I’m geeky, I love TED. But one dot that very much surprised me, closely related to me — the green dot of Quora, Quora the social Q&A network.

ideas, Gephi, "network visualization" "Quantified Curiosity" Amy Robinson

ideas, Gephi, Quora, "network visualization" "Quantified Curiosity" Amy Robinson, "Quantified Self"This [left] is a graph of Quora. It’s highly infused with all the different communities of my ideas.

These are beautiful graphs; they’re elegant and nice to look at but what do they mean? What can you actually learn from exploring ideas in this type of way?

It puts them into context. By being able to see my ideas and see how they’re connected to each other, I’m able to think about myself in new ways. I’m able to see, rather than just the fact that I started a new blog or I sent out a lifebonus email to friends, I can see how that evolve and where it came about. Based on the features of these graphs, I can actually understand more about where my ideas come from and how they change over time. And there’s a lot that can be done in Gephi that I haven’t even gotten to yet.

Really, like that one line at TED, those 5 seconds carried a much greater weight than just 5 seconds. I think the same can be true of ideas. How do I remember what was new to me four years ago? How do I understand how the ideas that i encounter today are influencing me as a function of time? And I really wonder how I can discover more ways to think about myself and how I can explore how my mind looks relative to other people’s. I wonder if there are hidden patterns inside of this.

I don’t know the answers to these questions but I think that there are answers, or can be. I’m very curious to understand who I am and how I exist. Consciousness is my greatest curiosity and in the end I’ve learned that we need to think socially about how to better think about thinking. This was a momentous task to put all this  together and it can certainly be done more efficiently. Remember, you are extraordinary. Your mind is exquisite. You, the things that you think about and the things that are important to you, create who you are and who you will become. So imagine how you might answer the question “what inspires you?” if you had a quantified mind in your cognitive toolkit.

Thank you.

ideas, Gephi, "network visualization" "Quantified Curiosity" Amy Robinson, Quora, beautiful, video, self, quantified mind

Continue Reading