on spiritual poverty

My engineer brain wants to cure all material scarcity.

While climbing a long staircase up a mountain to the Savitri Temple in the holy city of Pushkar I encountered a man whom I believe lived in the forest and collected firewood to sell to the villagers. Didn’t know if he was an ascetic or perhaps an untouchable. But I did conclude, without evidence beyond educated guess, that I make more money in a day than he makes in a year… or a lifetime.

I own X number of hats and Y pairs of shoes.

Suppose we did cure all material scarcity: Everyone is fed, housed, and clothed. Everyone has smart phones and computers and transportation to the farthest reaches of the planet. We can accomplish this with little additional technological innovation, and I think it a noble goal.

Then we’ll all want art and meaning. We’ll want to convince each other we are right. We might annihilate ourselves. We’ll create new scarcity with intellectual property—we might all be clothed but some of us will get haute couture—we’ll still use our plumage to create rank.

Stars upon thars… We might annihilate ourselves.

My communist soul wants a leveling revolution. My Christian soul wants the Golden Rule to underpin material society.

But we might annihilate ourselves in attempt to create that world. And I can’t argue with the fact that capitalism has lifted more people out of poverty than communism or religion combined.

My anarchist soul just wants to love God. No popes, no trips to Mecca. Just ecstatic prayer transmitted from the RV that I live in.

I own X number of hats and Y pairs of shoes. I’m a fashion legend in my own mind. I’m more creative than most. My IQ falls in the 98th percentile.

But God can raise a 1,000 of me out of the dust. I don’t mean much.

My favorite holiday is Dia de los Muertos; reminds me that rich or poor, we all snuff it in the end.

My engineer brain wants to cure all material scarcity. But does God really care about that? We are spiritually impoverished in the Global North. Spiritual malnutrition is destroying us.

Spiritual malnutrition is destroying us.

I don’t give a rat’s ass who you fuck, what you drive, or who you pray with. I just want you to talk with God.

Axis Evil featuring Napalm Fatale @ the Che Café – 24 January 2019

My performance with special guest Queen Mab playing the Kaossilator Pro. I’m using a seven-string guitar and trying new embellishments in the guitar parts, especially in the song “Voice in the Distance”:

lace punk, satin punk, petticoat punk, oh my!

This article was first published on the fashion blog Mad Hot and Artsy on 20 November 2018. Special thanks to them for featuring my content!

To satirize my proud (and adopted) feminine modality, I developed a personal clothing style I call “lace punk”, “satin punk”, and/or “petticoat punk”. It carries steampunk, burlesque, and pin-up influences; and emphasizes poise, class, and elegance. In other words, I give the appearance of “high class”. This combination kills when combined with sound posture and a confident stride—my satire has since evolved into a robust display of raw power.

Key elements involve millinery, corsetry, lace or satin gloves, heels, and frequent appearance in gowns.

The “punk” in all this is me: Accomplished hacker. Cyberpunk. Skilled guitar shredder. Free-thinking anarchist. These outfits tweak guys’ expectations when I talk engineering circles around them—“geek chic” never celebrated femininity quite like this.

I perform under the stage name “Napalm Fatale”. Have released two albums freely available at https://napalmfatale.bandcamp.com. Wrote an important article called “This is Transgender Music” describing this work and my musical goals.

I founded the company Whole-Systems Enterprises, Inc. to pay the bills. Am extremely interested in applying artificial intelligence to the fashion industry.

thrift store mysticism

I treat thrift store shopping as a mystical experience, as a spiritual discipline.

Being somewhat of a mystic, and a massive optimist (see my post “curvilinear optimism“), I tend to believe that the Universe (or the Divine if you prefer) provides what we need to accomplish our missions in life as we need it (or immediately before).

Today I went to Goodwill and found four perfectly-fitting ladies’ business suits. All match my design ethic of “obvious femininity”—i.e., they are not simply men’s styles adapted for women. All have skirts, because, as readers of this blog know, I refuse to wear pants. All were well-made and extremely inexpensive.

The occasion is timely: I’m preparing to meet regularly with potential investors in the startup I contract with, representing the technical side of the company (I currently serve indirectly as that company’s R&D leader). Therefore I need managerial-level business attire, and a lot of it.

My optimistic, mystical self interpreted this Goodwill shopping haul as a “sign” that I’m “ready” for the business responsibility coming my way.

Asserting the Feminine

I stressed above the “obvious femininity” of the outfits. Feminism in the 1970’s and 1980’s urged women in corporate America to “act more like men”. That ethic led to women’s suit designs that really just mimicked masculine designs. (Shoulder pads, anyone?).

But diminishing the feminine to advance in the business world only marginalizes femininity in general—and makes many women simply unhappy. The truth is, while gender definitely moves on a spectrum at individual resolution, as a whole we can argue that women differ from men. We can argue further that that difference can (and should) add just as much value to the corporate world as masculine traits do.

So I for one will only wear business attire that screams “feminine”. I will not mimic a man. And I’ve taken a hit in corporate America for doing so… but I don’t give a shit because I know women are the future of business (but that’s a whole different topic).

Part of this practice goes back to my early days of living as a woman, where I learned quickly that to be called “she” I had to wear extremely feminine attire. In other words, I had to donate all my t-shirts to Goodwill and stop wearing pants. Now that my face has been surgically modified, my voice is higher in pitch, and my hair is longer I no longer experience this issue. But my memory proves long…

Strange Effects

The corporation I hold majority shares in gives 10% of its income to secular charities. Goodwill Industries of San Diego receives most of it, and the cash donations are made through local stores. As a result, the staff of the North County Goodwill stores have come to know me, resulting in two unexpected effects:

First, recognizing that my personal style is almost entirely constructed from thrift store finds, they now seek my opinion on displays, which I am thrilled to give. It’s nice to be seen as a style authority!

Second, the women working in these stores have become familiar with the kinds of items I typically look for, so when I enter a store I can now find these women first-thing and ask for recommendations based on their knowledge of what has recently been placed on the racks. But they don’t just try to accommodate my style, they suggest their own ideas. This proves fun for everyone involved.

The money the business gives created this situation, but the fact that I’m simply nice to everybody nurtures it along.

See Also

curvilinear optimism

a corset and an LSD trip

I’ve dropped acid trice in my life, both times for religious reasons. The first time was on New Year’s Eve in 1993, and the second last Saturday.

During the latter experience I took detailed notes, which I may publish on this blog in the future. What can I say? I’m a scientist!

I respect the drug, so before each trip, I took prudent steps to ensure physical and psychological safety. So this last time I tripped while wearing an extremely tight-laced corset. You see, corsets make me feel cozy—like being constantly hugged. I wear them whenever I’m experiencing psychological distress, which is often. They comfort me. (Please don’t ask me to explain this right now—I know why, but I’ll save that matter for a later article).

I believed this decision helped keep me remain positively “grounded” during the trip; i.e., it limited the risk of a “bad” trip. I would have embraced a “bad” trip as educational, but I’m happy to report my experience went fabulously. Took advantage of LSD-driven thought processes to gain clarity regarding my love life and my frequent suicidal ideation.

More generally, I made sure to dress extremely feminine before proceeding. I expect this helped as well. Note the lace socks in the picture below. (Also note that I wore lace gloves during the trip which are not pictured below):

declaring myself an arbiter of proper ladylike behavior

Today I officially declared myself an arbiter of proper ladylike behavior. Issued the announcement via Twitter and Facebook:

Obviously I’m not a perfect lady myself, having immediately mocked the whole concept by using foul language in the second sentence of the announcement tweet.

But this speaks to a fundamental issue: A proper lady will not take herself too seriously! A proper lady knows that “ladylike behavior” is an abstraction and a ruse, yet chooses to employ it anyway. It’s a means to an end, and, for those that wish to participate, one of many pathways an individual may take toward creating a more civil and empowered society (if taken within proper context).

For a transgender lady such as myself, and perhaps for all ladies, ladylike behavior exercises empowerment. It provides an assertion of identity against a world that devalues the feminine. Deep in my transgender femme brain (and I’m only speaking for myself here), becoming a woman is never enough. I need to blossom into a lady. A “proper” lady. This liberates, not oppresses!

The best thing about this process is that I get to define “proper” and define “lady”. I’m creating something that works for me within the current time and place. Certainly I draw on a multitude of others’ etiquette manuals, blog posts, and how-to videos. I tap the Kama Sutra and the Bible for ideas. But the tiara stops with me—I’m the ultimate arbiter of my intent.

However, I plan to actively influence culture with my process and conclusions. Therefore I will add my voice to the growing worldwide call for promotion of civil, polite, feminine, demure (when appropriate), and of course, “ladylike” behavior by (interested) women of all ages. Will never treat my contribution as mandate, as many fabulous women will find no interest in it. This is perfectly fine.

My perspective proves unique in that no one taught me proper ladylike behavior growing up. The result is that I still “man-spread” and chug my beer when I lose mindfulness. As a work in progress recreating my own social construction from the ground up, I assimilate ladylike behavior as a foreigner learning a new language from scratch. This is beautiful and absurd. And it means all assumptions fell off the table.

So in my derivation of ladylike behavior for this social reconstruction I’m learning a lot about it, and intend to share my findings from a position only a transgender lady can offer.

Let’s get started!

Ladylike behavior involves many “musts”. I now issue my first:

“A proper lady never wears flip-flops in public, except at the beach, the pool, or the public shower.”

Proudly developed this “rule” myself; read it in no style guide or etiquette post.

The world is ours, ladies!

Update 27 April 2018

Received the following perspective-enhancing reply to my Facebook announcement:

Just reminds me to follow what I first admonished above:  “A proper lady will not take herself too seriously!”. Also illustrates how “expertise” lies in the eye of the beholder.

women’s style recommendation with artificial intelligence (part #2)

In “women’s style recommendation with artificial intelligence (part #1)”, I introduced my work toward developing artificial intelligence (AI) for fashion and style recommendation. Essentially, its an expert system built on a Bayesian belief network. Now I discuss model validation and next steps in the design iteration process.

I first wanted to see if the trained network correctly returned known recommendations (“wear” or “don’t wear”) based on known clothing selections. This procedure successfully validated the code I wrote. Then I wanted to see if the model can derive new style rules. Experienced partial success on this account; I will outline a possible strategy for improving it.

The rest of this article details the processes summarized in the previous paragraph:

Consider the following trained Bayesian belief network structure:

While calculating the structure, the learning algorithm also calculated the node value probability distributions from the training set:

We first evaluate the model on three fashion rules, asking whether the selected node combination’s values are okay to wear:

  • IF body shape = “apple” AND skirt zipper = “on front” THEN wear = “No!” [1]
  • IF body shape = “apple” AND skirt zipper = “on side” THEN wear = “Yes” [1]
  • IF shoes = “flip-flops” THEN wear = “No!” [2]

(I trained the model upon 126 such rules simultaneously).

Running the inference code:

All looks good. As a control, I added “shoe = pumps” (instead of flip-flops) to the above calculation, and see that these are okay to wear as expected. (However, see the discussion below where I ran into trouble).

So now I start to derive novel new style rules from the model. Suppose we want to simply find out if it is okay to “wear” an “apple” body shape. We expect the model to report “yes”, as it does, assigning a probability to the conclusion:

However, the model cannot handle the addition of a shoe type to the “apple” body shape query above:

The problem is “fixed” when I add a style rule specifically allowing apple-shaped folks to wear pumps, but I am not happy with this. Ideal outcome would be for the inference to conclude this. I’m first going to check the dependencies encoding… which, if that solves the problem, stresses the importance of specifying dependencies well in additional to lateral relationships. For example, I might establish a “human” node, and indicate that each clothing article and feature proves appropriate for humans to wear. Then I’ll declare that each body shape associates with “human = true”.

Nonetheless, the progress reported here is significant!

I’ll keep you posted.

– Emily

References

I figured out this whole “ladylike” thing today

“A ‘lady’ is a woman who, through her mere presence, simultaneously commands power while setting others at ease.” – Emily Marie Williams

I achieved public womanhood on 14 July 2015 when I declared myself a woman before a judge and started living full-time as one. But as discussed several times on this blog, I work consistently to develop my personal concept of “lady” (a concept distinct from “woman”), and labor to assimilate this concept’s traits into my core being. (Check out the “See Also” section at the end of this post for links to my previous writings on the subject, which illustrate my progress through this exercise).

Today I experienced a breakthrough in this concept’s development, upon writing the following letter to my mom. I quoted the key innovation at the introduction to this post:

I have discovered that when I dress simultaneously classy, elegant, and ladylike; and move with casual grace while wearing heels, strangers take me more seriously. They step out of the way in stores. They open doors for me more frequently. They resolve conflicts with me more effectively.

Of course, it helps that I smile at and make eye contact with everyone I pass, and that I’m tall. And that I’m confident in my skin. Somehow I’ve discovered how to command power while simultaneously setting people at ease.

I think that last sentence is the essence of the “ladylike” concept I am striving to create for myself. I now have a vision that fits my feminist ethos and still matches my extremely gendered ideas about class.

Recently concluded that my days proceed more effectively, both in my mind and out in society, when I dress sharply.

Here is what I was wearing when I figured this out, what earned me the respect from strangers I received today that enabled me to put the pieces together:

See Also

women’s style recommendation with artificial intelligence (part #1)

Introduction

We know several basic style “rules” (ha!) based on body shape:

  • Skirts:
    • “Apple” Body Shape:
      • IF body shape is apple AND skirt has front zipper THEN don’t wear
      • IF body shape is apple AND skirt has side zipper THEN wear
      • IF body shape is apple AND skirt has no zipper THEN wear
    • “Rectangular” Body Shape:
      • IF body shape is rectangle AND skirt has front zipper THEN wear
      • IF body shape is rectangle AND skirt has front zipper THEN wear
      • IF body shape is rectangle AND skirt has front zipper THEN wear
      • IF body shape is rectangle AND skirt is A-line THEN wear
  • Pants:
    • “Apple” Body Shape:
      • IF body shape is apple AND jeans have flare THEN wear
      • IF body shape is apple AND jeans have pleats THEN don’t wear
      • IF body shape is apple AND jeans have stretch THEN wear
      • IF body shape is apple AND trousers have flare THEN wear
      • IF body shape is apple AND trousers have pleats THEN don’t wear
      • IF body shape is apple AND trousers have stretch THEN wear
    • “Rectangle” Body Shape:
      • IF body shape is rectangle AND jeans have flare THEN wear
      • IF body shape is rectangle AND jeans have pleats THEN wear
      • IF body shape is rectangle AND jeans have stretch THEN wear
      • IF body shape is rectangle AND trousers have flare THEN wear
      • IF body shape is rectangle AND trousers have pleats THEN wear
      • IF body shape is rectangle AND trousers have stretch THEN wear

We want to create an artificially intelligent system to probabilistically decide, given a query such as “I have an ‘apple’ body shape and am thinking of wearing a skirt with a zipper in front. Should I?”. To accomplish this we use these rules to train a Bayesian network, and then use the network to make inferences upon queries such as the one given above.

Training the Network

From these we derive the 13 nodes of our Bayesian network:

Node
apple
jeans.with.flare
jeans.with.pleats
jeans.with.stretch
rectangle
skirt.with.a.line
skirt.with.front.zipper
skirt.with.no.zipper
skirt.with.side.zipper
trousers.with.flare
trousers.with.pleats
trousers.with.stretch
wear

We use the rules and the nodes to produce an automatically generated graph. Put to help it along, we will apply some expert knowledge and specify some

We seed the model structure identification algorithm with some basic expert knowledge by manually specifying the following 12 causal relationships:

From To
rectangle wear
apple wear
skirt.with.front.zipper wear
skirt.with.side.zipper wear
skirt.with.no.zipper wear
skirt.with.a.line wear
jeans.with.flare wear
jeans.with.stretch wear
jeans.with.pleats wear
trousers.with.flare wear
trousers.with.stretch wear
trousers.with.pleats wear

(We will see later that the automated graph structure learning procedure adds one more edge).

We save these relationships in “output/style_edges.csv” for later import using R.

We then encode the rules in dictionaries/hashes for items co-joint in a rule. For example, we express the skirt-related rules pertaining to apple-shaped bodies in JSON as:

    {
        "wear": "Yes",
        "apple": "1",
        "skirt.with.no.zipper": "1"
    },
    {
        "wear": "Yes",
        "apple": "1",
        "skirt.with.side.zipper": "1",
    },
    {
        "wear": "No",
        "apple": "1",
        "skirt.with.front.zipper": "1",
    }

For each entry, we zero out all other nodes (expect for “wear”, which is set to “No”), and express all 19 rules as a data frame, where the index order corresponds to the node order displayed above:

0,0,0,0,1,0,0,1,0,0,0,0,Yes
0,0,0,0,1,0,1,0,0,0,0,0,Yes
0,0,0,0,1,0,0,0,1,0,0,0,Yes
0,0,0,0,1,1,0,0,0,0,0,0,Yes
0,1,0,0,1,0,0,0,0,0,0,0,Yes
0,0,0,1,1,0,0,0,0,0,0,0,Yes
0,0,1,0,1,0,0,0,0,0,0,0,Yes
0,0,0,0,1,0,0,0,0,1,0,0,Yes
0,0,0,0,1,0,0,0,0,0,0,1,Yes
0,0,0,0,1,0,0,0,0,0,1,0,Yes
1,0,0,0,0,0,0,1,0,0,0,0,Yes
1,0,0,0,0,0,0,0,1,0,0,0,Yes
1,0,0,0,0,0,1,0,0,0,0,0,No
1,1,0,0,0,0,0,0,0,0,0,0,Yes
1,0,0,1,0,0,0,0,0,0,0,0,Yes
1,0,1,0,0,0,0,0,0,0,0,0,No
1,0,0,0,0,0,0,0,0,1,0,0,Yes
1,0,0,0,0,0,0,0,0,0,0,1,Yes
1,0,0,0,0,0,0,0,0,0,1,0,No

We save this data frame as “output/style_rules.csv” for later import by R.

In R, we load the necessary libraries and the CSV files. We also ensure everything is a factor in the rules data frame:

We look at the expert-specified edges, noting the existence of 12 relationships. After running the hill climbing algorithm to derive the network structure from the prior-specified edges and the rules, we notice that now 13 edges are present:

Here is the added edge:

From To
apple rectangle

We derive the model’s parameters from the training data, and then compile it for use in inference.

Results

Suppose we have an “apple” body shape, and want to choose a skirt using this model. We try the following skirt types against the apple body shape to infer whether or not to wear a particular skirt:

The first result in the image above resoundingly rejects wearing a skirt having a front zipper when one carries and apple-shaped body. By contrast, the second result approves of skirts having side zippers for apple-shaped folks. Both results concord with the IF-THEN-ELSE rules initially specified. The third result proves interesting—we did not provide a rule for apple-shaped bodies and A-line skirts, so the model provides no conclusion.

We observe similar results for trousers: The first two outcomes match the rules, but the third provides no decision because we provided no information about whether flare and stretch may be used together in a pair of trousers for apple-shaped bodies, or for any body shape for that matter!

Issues to Resolve

As indicated in the last paragraph, in practice a pair of trousers may have both flare and the ability to stretch. Each of these traits alone proves great for apple-shaped individuals. So together I manually infer that the two together are at least okay and may be even preferable. However, the model does not derive such a conclusion. In other words, we need to add rules saying these two traits may coexist.

Also, this effort took a lot of manual “expert” specification of the initial “seed” graph structure. Ideally one would learn the final structure purely from rules. My thinking is that the rule data frame is rather sparse, making it hard to learn the structure in an automated fashion. On the other hand, I may not have chosen the best learning algorithm.

Stay tuned…

– Emily

Update 16 April 2018

I’m onto the next iteration of the model design. A visual of results so far:

why I didn’t suicide this morning

I’m publishing this strategy because maybe it will help someone else survive in the future:

I feel an intense psychiatric compulsion to suicide every time I experience romantic heartbreak. It’s simply part of my bipolar disorder (which I improve management of everyday). Usually I’m well adept at handling these situations and moving quickly out of them.

However, this morning’s heartbreak incident proved much more difficult to manage; I moved beyond psychiatric compulsion and actually considered offing myself. It’s not that the woman involved is any more awesome than previous situations (she is), it’s just that I feel beaten down from years of serial heartbreak. I always get back up again after getting knocked down. Won’t ever stop doing that. But sometimes one just needs a break.

To survive, I constructed the following two-part argument, built entirely on the deep love I have for this woman. I reframed survival as an expression of this love:

First, and less importantly, she asked me not to contact her in the future, which I intend to honor. I realized that if I took my life she would certainly find out—which be a form of contact—a message somewhere between “I love you” and “fuck you”.

This first reason isn’t particularly rational, but the second reason carries tremendous clarity and precision:

Taking my life in response to her rejection would likely traumatize her, even though my action wouldn’t be her fault by any stretch of the imagination! I could not do this to her. She’d spend a lifetime second-guessing her decisions, and potentially years on unnecessary guilt. She’d perhaps require therapeutic intervention. It would disrupt her life in significant ways. I cannot do this to someone I love, who’s well-being I care so much about. So better for me to endure the pain which will pass in time.

The interesting thing about this last reason is that, while I constructed the argument expressed above based on concern for her well-being, I was even more concerned about her son’s well-being (who doesn’t know me). Here is my logic: If my beloved withdrew into trauma due to a suicide on my part, she would be less able to provide emotionally for her son. This of course would do great damage. Again, I cannot do this to someone I care about, so I’ll endure the pain.

So instead of killing myself I went shopping.

mathematical coolhunting

I aim to become the Timothy Leary of data scientists!

Intuitive coolhunting scales poorly. Here’s some math to help fix that problem:

Axioms of cool

Five axioms enable us to mathematically model cool:

  1. No one is intrinsically cool, individuals simply channel it.
  2. Ability to temporarily hold coolness varies by individual.
  3. Coolness naturally flows into some individuals more readily than others.
  4. Rate of coolness flow into an individual increases with the amount of cool stored within that individual’s social network.
  5. The rate at which cool leaves an individual increases as observation of cool’s presence in that individual increases.

Examining these axioms in more detail:

1. No one is intrinsically cool, individuals simply channel it

‘Cool’ flows into and out of individuals, as shown by the following stock and flow diagram:

Individuals can temporarily store some of this cool, in a manner resembling a capacitor storing electrical charge. We can for example imagine an individual’s step response to incoming cool:

We can describe this capacitive behavior in the stock and flow diagram with first-order dynamics:

2. Ability to temporarily hold coolness varies by individual

Individual capacity for storing cool differs. Given the same step input above, we might observe different responses for individuals A and B:

3. Coolness naturally flows into some individuals more readily than others

Some individuals channel cool better than others. We model this by varying the “natural” coolness input flow rate by individual:

4. Rate of coolness flow into an individual increases with the amount of cool stored within that individual’s social network

Individuals with cool friends tend to more successfully channel cool themselves. We model this by increasing influx rate according to a “coolness in social network” factor:

5. The rate at which cool leaves an individual increases as observation of cool’s presence in the individual increases

Once observed, cool tends to exit the individual it was observed in. We model this by increasing the coolness decay rate as a function of public observance of an individual’s coolness:

Source and sink of cool

Assume the universe provides an infinite source of cool. Similarly, assume existence of an infinite capacity sink for coolness that exits individuals. Also assume that everyone alive connects to this source and sink. It follows that individuals cannot “use up” the supply of cool or withhold coolness from others. Under the axioms, cool never transfers from one person to another—the relationships between individuals simply modulate the rate cool enters each person from the source and leaves each person to the sink.

Networks of cool

The last two axioms relate individual ability to receive and store coolness to the instantaneous state of their social network. To demonstrate the axioms in this social context, suppose the following friendship network exists among seven individuals:

Now suppose that Julie holds a lot of cool at a particular moment. It follows from axiom #4 that Guido’s instantaneous ability to channel cool will increase due to his connection with Julie. Similarly, if Di stores very little cool at a given time, Hardeep’s ability to receive cool will not benefit from his relationship with Di.

Hardeep’s coolness influx rate benefits from the combined cool stored within Emilio, Kaitlin, Di, and Abe. However, because of axiom #5, the fact that Emilio, Kaitlin, Di, and Abe observe Hardeep’s cool accelerates its exit from Hardeep. Due to the first-order dynamics described above, this exit of cool lags the influx of cool, giving Hardeep time to enjoy a temporary build up of coolness and time for Emilio, Kaitlin, Di, and Abe to benefit from its presence in Hardeep.

Simulating coolness networks

Using the mathematical framework developed above, we now simulate cool’s flow within the network described in the last section. Since we currently have no way to actually measure cool—and therefore parameterize the model—we run it with fictional initial conditions and examine the resulting system-level effects to see what happens.

The combined model for this friendship network is shown in the image below (sorry about the mess of arrows):

Simulating this model with arbitrarily selected initial conditions and factors yields:

A long way to go before this work is useful

As stated above, we currently have no way to measure cool, and therefore no way to validate and parameterize this model. Expect a Bayesian strategy to emerge shortly though. Until then, this work remains conjectural and exploratory.

Computation notes

Used Vensim PLE to draw and simulate the stock and flow systems, R to display the simulation output, and NetworkX to draw the example social network.

encoding fashion rules into mathematical data structures (part one)

As we build our fashion recommendation engine, we seek rules to populate it with. With few exceptions (e.g. [1]), we find these rules encoded in prose or infographic form, rather than a semantic web form suitable for computation. For example, [2] provides written advice on dressing fabulously for a “rectangular” women’s body type. The writers meant this document for a human reader, not a computer program.

However, we can’t scale a process consisting of manual extraction of rules to the level we would like to achieve in this project, so we turn to natural language processing to extract rules from texts in an automated fashion. We begin by identifying parts of speech and the syntax relationships between words in sentences. For example, consider the following two fashion rules from [2]:

  • If you are a heavy or tall rectangle, choose a big bag.
  • If you are a petite rectangle, choose a petite bag.

We then create a directed graph with words as nodes, each with an attribute indicating its part of speech, and edges indicating the syntactic relationships between the nodes (e.g., “heavy” is a modifier of “rectangle”). We also add edges to specify the direction of sentence flow. Visualizing the above two sentences in this form using Neo4j [3] yields:

Next Steps

In the next phase, we plan to automatically derive computationally useful IFTHENELSE rules from such mappings. For example, the above two sentences express in IFTHENELSE form as:

  • IF rectangle AND (heavy OR tall) THEN choose a big bag
  • IF rectangle AND petite THEN choose a petite bag

Once we form a comprehensive set of such rules, we will load them into an expert system or related system to enable fuzzy reasoning on the rules, enabling custom fashion recommendations!

After this, we will come up with a way to reconcile similar recommendations. For example, suppose we find the following two IFTHENELSE rules from two different sources:

  • IF rectangle AND (heavy OR tall) THEN choose a big bag
  • IF rectangle AND (heavy set OR tall) THEN select a big bag

These say the same thing. We will devise a way to combine them into one recommendation such that the weight (value) of the recommendation doubles due to its backing by two distinct sources.



References

  1. Vogiatzis, D. Pierrakos, G. Paliouras, S. Jenkyn-Jones, B.J.H.H.A. Possen, Expert and community based style advice, Expert Systems with Applications, Volume 39, Issue 12, 2012, Pages 10647-10655, ISSN 0957-4174, https://doi.org/10.1016/j.eswa.2012.02.178. (http://www.sciencedirect.com/science/article/pii/S0957417412004411) Keywords: Style advice; Recommender system; Fashion ontology; User modeling
  2. http://www.styled247.com/rectangle-body-shape
  3. https://neo4j.com

celebrating myself this Valentine’s Day

No significant other shares their life with me, a fact that usually annoys me on Valentine’s Day. This year I decided not to fuss about it, choosing instead to celebrate my healthy love for myself. So I’m wearing cute Valentine’s Day lingerie for my own joy.

Girls, if you find yourself single next Valentine’s Day, get yourself some sexy lingerie and delight in your own company!