Couple of minutes pass by, Baba goes to Ma to ask “Where’s Ridwan?” assuming I made it to my destination. She bewildering looks at him and goes “I thought he was with you?…” and then the fear sinks in. They both realized they lost their 3 year old bundle of joy.

Baba starts scampering, Ma’s worried sick, she doesn’t know what to do, and I’m — well I’m probably somewhere annoyed lol Then some guy notices Baba is pretty frazzled and looking for a child and so he tells him in Hindi “Oh, I think they went this way!” … why he would say that? 🤔 But at that point, proud parenting moment incoming, Baba activated all his spidey senses and realized the guy was probably lying so he went, THE OPPOSITE WAY! YES, GO BABA GO!

He activates Usain Bolt mode and starts sprinting in the opposite direction of where the guy initially tells him. Then he catches up to ANOTHER guy, who somehow…has me in his arms (the f***? explains why I’m probably irritated) That guy starts playing dumb and starts saying in Hindi “Oh, who’s bachcha is this? Who’s bachcha is this?” Because somehow, I ended up a hundred yards away, with my two little feet, and into this man’s arms … the haaaaill??

Baba at this point is probably just relieved he got his son back and we regroup with Ma and leave. Moral of the story, don’t let your kid out in a busy shopping center expecting him to arrive in their mother’s arms. He has two little feet.

Alas, something I’ve thought about for a while. The Apple Hivemind. It’s not that I hate Apple (I promise), it’s just when their objective is to scream at everyone through their captivating marketing campaigns that they’re the best, well — you might want to question their need to do that.

If you recall, back when Steve Jobs was alive (Rest In Peace), I truly believe they were — the best. The best consumer technology company. They had the best phone, tablet, personal laptops, and computers. It was AMAZING! I remember how excited I’d get for every Steve Jobs launch. Not that I actually watched it (lol), but I knew, this man was about to pull back the curtain on the latest and greatest piece of technology — it was mesmerizing! I mean, how could you not get so excited as a teenager to know that the iPod was now going to be in color! Yes, folks remember that?! Long gone were the days of those unlively black and grey pixels before Kindle made it cool, now we could use the full RGB spectrum! Amazing! Or remember when the MacBook went from the pasty white to the cool, metallic grey?! Holy shit! It was like something from the future.

You know what really blew everyone’s minds? Back in 2007, there was chatter about a touch-screen phone. First off, what the f*** is touchscreen?! This ain’t no Tom Cruise in Minority Report (if you get the joke, it means your past your prime). But seriously, touchscreen?! I remember we were all talking about it in Mr. Mata’s Music Class. I remember my boy from back in the day, Nick was really leading this convo (or maybe it was Cam? Okay whatever it was one of them). I remember sitting there completely spellbound wondering “I can’t even imagine this?” Then we had it, the unveiling of the brand, new iPHONE!

It was everything we had imagined and MORE! Jobs had done it again! He had us guessing and delivered like… honestly, I can’t even make a comparison because it would be an insult. He had mesmerized us — we were completely captivated! This, ladies and gentlemen, WAS IT! This was the pinnacle of personal, handheld technology and we all needed to have one! Syke! My brown parents ain’t buying me that. My white T-Mobile flip phone was good enough. Iykyk. Those were truly the glory, no, golden days of Apple.

But with everything, nothing is beyond the boundaries of nature — whatever goes up must come down (Yeah, yeah cheesy but true). Eventually he passed and I noticed a shift in the releases and products. I think it carried on for a little bit but I don’t recall that same magic as before. Tim Cook seemed like he did his best or honestly lack there of but eventually once apple became a trillion dollar company it was pretty obvious — Apple just wanted the cash. Not that they weren’t like that before but at least the cash I used to buy their products with felt worth the value or at least Jobs made me believe it was. Now, to me at least, it’s obvious there’s a lack of innovation.

Every time an iPhone is released I’m wondering “Alright, what’s so special about this one besides the price tag?” Even the brand new MacBook I’m typing on right now has a lot of similarity to my MacBook — from 2011! When Jobs was around there was none of that. You never thought of the past because he was literally leading us into the future. No, seriously he was. I looked up to everything he did in the ladder half of his career. I even have his saying on my childhood bedroom door “The people who are crazy enough to think they can change the world are the ones who do” Yeah, I actually believe that. And that’s what I mean, that desire for innovation, how he made us believe is gone …

Now Apple is left with clinging to their iMessage and FaceTime apps and gaslighting Android users that their inferior (Oh yeah, now it’s going to get personal). That individual up above mentioned how they’ve been left out of group chats because of their Green Bubbles. Yes! Left out. Do you know how many memories and connections are created in group chats? We literally laughed about our shared trauma over this and I know if you’re reading this some of you have felt the same. How is it that a cool, tech company now forces you to use their product or else you’re not relevant…to your own community. Bruh.

Honestly, it kinda sense why Apple is so exclusive. In Jobs’ biography, which I only read the first 160 pages of (lol), he mentions multiple times that essentially you’re either with Apple or against them. You’re either with us or against! That was his younger self but it seems that transcended onto the later years of Apple. I’ve always wondered why that was the case? Maybe because he has trauma from his adoption from when he was initially returned. Yes, his first adopted parents returned him as a baby before being eventually adopted by his loving parents. Maybe he did everything to hold onto what he had or considered family. Who knows but alas, I don’t want to speculate too much over someone that’s passed.

Certain tech in Android, specifically my Samsung, have been around for years before Apple even had that twinkle in their eye. For example something my brother showed me is if I hook up a monitor to my Samsung S21 (from almost 4 years ago), it operates as a fully, functional PC. Yes, it literally turns into a computer. Then you connect a bluetooth keyboard and mouse and you have a fully-functioning work station. You know widgets? They have been around on Android for as long as I can remember. Same goes for the fingerprint reader, better cameras, so on and so forth. See, if Apple were to lead the charge on these and actually introduce them, then we’d have that magic and innovation back. But they don’t and it’s just sad.

Overall, I’m sad about Apple’s current state. It’s sad because this was once the company that literally led us to innovation and now, they cling onto the past and force their users (including myself as a MacBook user) to believe that what they create is amazing. It’s not … it’s just seeing what’s the most popular feature out there, implementing it years later, bumping up the price, and creating an illusion that if you don’t buy this you’re a loser. It’s elite capitalism. And you know what, we’ll all do it because we don’t know any better …

I wish companies created to create and not just to make money. I understand we live in a capitalist society and money is a huge part of it but creativity is important. You don’t want to look back on life and show how much you had in your 401k but rather what you were able to create from your own unique thoughts.

At the end of the day, I’ll remind myself Apple may be a shell of its former self but at least we were able to witness it — that grandeur, those captivating moments. People think I’m angry at Apple, no, I’m just sad. Sad that we saw it at its peak and now we’re watching it come down. Steve Jobs was that guy for us that really made us believe we could do anything because we were literally watching him do it. At least we had someone to show us it was possible. Thanks for being just enough crazy and being our Tony Stark.

I want to start a series where I share my plane stories I have with the passengers next to me. Back in the day, I used to fly a lot and I would always ask the person next to me “Going home?” to start off a friendly conversation (no I wasn’t trying to hit on them, I was generally curious). However, you would have no idea the success rate that it had starting an actual hours long conversation. The things you learn about these complete random strangers that I may never see again was just, wild. Not that they were telling me crazy things (okay, maybe some of them were crazy) but the fact that we are complete random strangers that just happened to cross paths at that exact time and we’re now connecting for those few hours, it was incredible. We would talk about anything and everything: life, background, people we knew (yes, one person and I had a mutual connection), occupations, where they’re going (obviously), and above all, family would almost, unequivocally be brought in one way or another. Whether it was “I have one brother” to “Let me tell you about my family background”; it was so fascinating. I’ve even managed a couple of rides out of these conversations LOL!

Anyways, I just wanted to get them out of my head and onto a medium (Ha!) where I can share them with everyone. So here goes nothing and my first story.

This one will probably be relatively quick. I remember I was flying back to San Diego, and I was speaking to a guy across the aisle. Somehow we started talking (I think it was different from the usual “Going home?” line) and I had been telling him how I travel/wanted to. That was also around the time Game of Thrones was hot, and I mean hot, like smoldering. He mentioned a place named Dubrovnik, which I hadn’t heard of at the time so I asked “Where is that?” He told me it was in Croatia and proceeded to ask me Have you seen Game of Thrones?” Ummm… obviously! Who hasn’t?! It was only the greatest TV masterpiece of our generation. Well, until Season 8 unfortunately …

Anyways, he goes onto explain that some of scenes (I think Kings Landing) were shot in Dubrovnik and that I had to to visit. I’m sure I obliged and thought “Man, I gotta go there” Sure enough I have yet to visit haha.

Regardless, it was a nice chat with this gentleman (can’t recall his name), and at one point he ended up asking me how I was getting home (and I’m sure at another point I had told him I lived in PB (Pacific Beach). YEEAAHHH!) So I’m sure I said Uber/Lyft and he said “Nah man, you can come with us” (I don’t think he said that exactly but go with it. It’s for theatrical effect 😂), and he told me his wife was picking him up. Of course I’m sure I was thinking/asked “uhhh you sure” (not knowing him that well really at all, and also I’m just going to show up to his wife’s car and be like “Hi I’m Ridwan, can I get a ride?” LOL like what the?) I guess these sort of things can only happen in particular cities in which this case would be San Diego (this actually reminds me of another story where 2 guys and a dog popped into my UHaul and one of them tried to sell me Diamonds or Weed. I’m sure you can guess what I took? The DIAMONDS! DUH! Annnddd they were obviously fake … )

Anyways back to the story, we pop out onto the curb and walk up to his wife’s car. He says “Hey this guy is going to get a ride with us?” to what obviously you can imagine his wife was a little taken aback. Husband shows up, I’m assuming she’s happy to see him, but confused/probably scared why he is bringing some random guy into their car. I’m just chuckling thinking about it 😂. Anyways, I get in the car and they take me to my apartment in PB. I say bye and that’s it.

So that was a bit longer than a I intended it to be but as I kept writing, certain details kept coming back to me. Anyways, hope you enjoyed this (whoever you are lol) and I hope I continue to write these and you keep coming back and reading them. I’m trying to do my best Dumbledore’s Pensieve impression here. Tootles!

Also, enjoy this last image of Dumbledore and Harry Potter just because I think it’s funny 😂 Okay, but seriously, later!

I miss you. I hope you’re doing well wherever you are. When you passed that hit me differently in a way I still can’t explain to this day. Seeing you suffer that way and at the same time knowing that this was probably the end…I just, I’m not sure. It’s hard to explain.

I’ve believed for a long time I can/will do anything I want to. Whether it’s achieving a personal goal, creating a new relationship, traveling and going somewhere, I’ve always been able to do it as many of us have been. But it was at that moment I realized…this wasn’t something I could fix. That this was the end. It was the end of your life and there was nothing we could’ve done to fix it. All we could do is stand there and accept it. Accept the fact that you wouldn’t be a phone call or even a plane ride away. Accept that when I left Bangladesh September 2018, that would be the last time I ever saw you.

It’s like a part of me went blank that day when Ma called and she said “Bhaiya is no more…” and she cried. She cried. My mother who I’ve always looked up to (yes, yet our differences) showed her vulnerability and there was nothing I could do to fix it. There wasn’t any bit of effort, comforting, ideation, nothing could bring you back. You had left and weren’t coming back…that was it.

At that moment, I didn’t know what to do. I think before all your children I had accepted that fate, that you would be gone and I think you had too. Even as I write this, those feelings are coming back. Saddest day of my life …

I’m not sure what else to say or even why I’m writing this. I just heard this song “With or Without You” by U2 on a show I was watching and it was playing during a funeral. First thought I had was you. You’re the closest person I’ve lost to date and I think…I think I’m still having a hard time accepting it. That you’re not there…

I plan on going back to Bangladesh this December/January with my wife, Humma, and I plan on going to your grave. I want to see you see where you were laid to rest. Even if it may be a plot of land I know that you were/are there. That’s why we have cemeteries right? To pay our respect to the ones past; to symbolize them and let their souls shine eternally.

Not sure what else to say but that I miss you, I love you, and wherever you are, I hope you’re at peace. See ya again, Nanabhai.

You’ll always be with us.

Disclaimer: This is my story and everyone’s experience with OCD or any other mental health condition will be different. The objective is to share my side in hopes that you feel that you’re not alone and maybe one day, you are encouraged to share your story with the world.

I’ve been wanting to share this for some time, but I wasn’t sure how to go about it. I figured writing an article sharing my feelings may be the best way. I’m not sure who is reading this but please, read with an open mind as it is already difficult for many of us to share our inner thoughts and feelings, and I am no different. We all have daily challenges which have been exacerbated by the pandemic or whatever else we may be going through and this is a glimpse into my journey. Its intention isn’t to be a sob story but A) to express what I’ve been holding in so I feel better. My mental health is my responsibility and no one else’s. I would love everyone’s support but I am responsible for taking action, and B) to let others know that they’re not alone and I hope one day you will feel comfortable sharing your story.

Alright now that that’s out of the way, phew, let’s begin.

I can remember my OCD (Obsessive Compulsive Disorder) since I was a child. I remember organizing all of my toys into rows and columns, folding my clothes and putting them onto the shelves, etc. Mind you, I was less than 10 years old at this point. It was just something in me. So far, this is starting off innocent and cute, great start!

Then as I grew older and my daily anxiety increased (OCD falls underneath the umbrella of Anxiety), and things started taking a more drastic turn. I started washing my hands more, like I felt there were germs everywhere, LIKE EVERYWHERE! This may still seem fairly “Okay, it’s not that bad” of a moment, and honestly, I didn’t think it was either, at the time. It felt like it was just a part of my routine: go inside the bathroom, do my business, wash my hands. HOWEVER, now I couldn’t touch anything else with those washed hands so then I’d put my elbows together to turn off the faucet and open the door knob. Thank goodness I didn’t feel the need to wash my elbows or else then we’d be in a doozy. Har-har-har…it’s a joke don’t worry.

I continued this throughout high school and then things got really interesting in college, that’s when the fun began. The OCD decided to finally infiltrate into the mother land, my thoughts.

I’m sure people might be sitting there thinking “Yeah, I have unwanted thoughts all the time, it’s not a big deal.” So with someone with OCD, it’s harder because we can’t let it go. Like we become (as the condition states) OBSESSED with that thought that we literally can’t let it go. It’s like the thought has taken over like Venom.

It’s like you have something else taking over your mind, and the more you focus on that thought the more it takes over. You let it define you — you let it take over your mental energy. It dictates everything you do throughout the day. You can have unwanted and ruminating thoughts about your feelings for someone, your sexual orientation, if you want to hurt someone, if someone is talking about you; the list goes on and on. And yes, these thoughts can get graphic. To the point where, you sit there and you question your character. You try your best to let go of these thoughts but you can’t — you just can’t because you’re in too deep. And then the worst part is, if these thoughts come to fruition then what, what do you do then? Now you’re really in for it, now it’s over — now it’s escaped your head and made it into the physical world, and that’s the fear we live in. What if I do do something that I regret? Again, I’m sure you may be sitting there suggesting “It’s just a thought, just let it go” Again, not easy. Not in the slightest.

During this journey, I’ve met individuals who are afraid to go outside of their own room because they believe they will be infected with germs and die. This fear is so strong that they confine themselves to their room and will not leave. When I heard their stories, it hit me like a ton of bricks. These thoughts have taken over so much that they cannot even get on with their daily lives. They are a prisoner of their own mind with no sentencing date. It feels like eternity.

I remember those days — I remember feeling like I couldn’t get past the current day. I remember all of that. I remember thinking “Man, let me just get through today, please.” You do everything in your power just to survive. You pray, you talk to your partner, you exercise, you go to the beach, but nothing, nothing gets you out of that — that prison. The worst part is, you feel like people don’t understand because especially, you don’t even understand so how can you expect anyone else to? Remember at the beginning of this I had said one’s mental health is one’s responsibility, well again this was mine.

You feel like a toddler again, not being able to articulate what you want to communicate. But at this point you’re an adult and you still can’t do it which seems so much more debilitating. You’re basically at a loss of what to do. Your anxiety at this point has heightened, which means more food for your OCD to feed off of and keep going. You feel like there’s no way out, and your stuck …

Then one day, you’ve fallen into this black hole and you feel like there’s no escape. You think “Man, is this it? Am I going to end this?”, and previously you thought your mind couldn’t go to a darker place but man, were you wrong. If you were in a mental prison before now you’re just in sheer darkness with literally no hope.

I definitely remember specific days where I contemplated suicide. It’s hard for me to convey how that feels, but for me there were multiple times where I had said “I’m going to kill myself”

1. It was for attention. Yes, as twisted as this may sound it is a thing and I’m sure many have felt that. Regardless, this isn’t good.
2. You’re actually contemplating it and wondering if this was it.

I can say, with 100% certainty, the second, at least for myself, was significantly worse. At least for the first, there was that glimmer of “I’m not going to do this. I just really need attention right now because I’m going batshit crazy, and I don’t feel heard, and I’m acting out, and I don’t know what to do.” The ladder — man, I can’t even begin to describe to you how that was. I felt myself slipping…

I honestly felt like it was the end.

However I made it, and I’m still here today.

I am still taking it day by day and continuing to work on my mental health. One thing to keep in mind, it is a marathon not a sprint, and some days will be better than others. The days where I’m down, I try to focusing on what I need and what will be best for me. The days I’m up, I try to take advantage of it and do as much as I can. Does it always work out that way? No, but I’m trying — always trying to climb up.

I hope this encourages anyone reading this to take their mental health more seriously and to give that kid inside some extra TLC when they need it. Also, I hope you will support me while I continue my journey because honestly — that would be a nice feeling to have; to feel supported, and of course I will support you as well.

Thanks for taking the time out of your day to read this. I really appreciate every single one of you. Much love to you all ❤

As a believer in Bitcoin since 2017, I bought into that sentiment or “hype” and never let go. As I’m sure many of you are wondering why I didn’t just sell at that time, it is because I believe Bitcoin will drive the future of the monetary system and force a change in the current one. However, for the time being, the question still begs, what drives the current price of Bitcoin?

Lets first begin with a quick background on how I exactly landed at this conjecture of analyzing the correlation between Tweets and Bitcoin. I started trading Bitcoin along with other cryptocurrencies after the 2017 epic pump to the colossal dumpster fire we experienced shortly after. Why, do you ask? Because I was driven by greed and the lust for eternal wealth. Eh-hm, after day trading and swing trading for nearly a year, which entailed learning about the different indicators such as Relative Strength Index, MACD, Oscillators, and etc. I realized I was a s*** trader. By that point I had already bought into Bitcoin and I knew it was a long term play for myself. I ended up adopting another strategy, Dollar Cost Averaging. I’d recommend Googling it if you are unfamiliar with the term as it may be one of the best financial terms you’ll learn. The concept being, I bought into Bitcoin a little bit every month and regardless of the stark fluctuations, and ideally my investments should come out in the green in the long run. Eventually, however, I started thinking and I finally, scientifically, arrived at the below conclusion:

The Price of Bitcoin is Driven By the People’s Sentiment

During my data science bootcamp I was contemplating how to combine NLP (Natural Language Processing) and my love for Bitcoin. Ah-ha! To Twitter!

The question at hand: Can sentiment from Tweets impact the price of Bitcoin. Please bare in mind, I could’ve chosen any form of electronic communication such as Telegram and WhatsApp messages, Facebook posts, and etc. but I decided on Twitter.

Next step, I formulated a plan:

1. Scrape Tweets from Twitter from two different time periods to compare.
2. Determine the positive/negative sentiment scores for these sets of Tweets.
3. Compare the sentiment with the prices over that given time period.
4. Determine my conclusions.
5. Build an app to exhibit to the public.

It was a pretty simple and straightforward plan. So let’s get to work!

Twitter has an API called Tweepy…

…which you can use to access their Tweets. First, I needed to apply for a developer account. I went to Twitter, created the developer account, and I was approved within 24 hours, voila! This could be longer for you as it will vary. With the developer account I had the keys I was looking for:

consumer_key = os.environ.get(‘CONSUMER_KEY’)

consumer_secret = os.environ.get(‘CONSUMER_SECRET’)

access_token = os.environ.get(‘ACCESS_TOKEN’)

access_token_secret = os.environ.get(‘ACCESS_SECRET’)

However, there is a limitation of how many Tweets you can scrape at any one time. This was a bottleneck in my process but luckily I was introduced to SNScrape. All I needed to do was call SNScrape with the given parameters, as shown below:

!snscrape twitter-search “#Bitcoin since:2020–11–01 until: 2020–11–02” > btc_tweets_daily.txt

[1] J. Dinco, Downloading Tweets with Tweet IDs via SNScrape & Tweepy (2020), https://medium.com/@jcldinco/downloading-historical-tweets-using-tweet-ids-via-snscrape-and-tweepy-5f4ecbf19032

I specified SNScrape to search Twitter for links to Tweets with the hashtag “Bitcoin” from the end of October 2020 and put it into a text file. All I have done with SNScrape up to this point is pulled the links to the Tweets as aforementioned.

Once I had the links, I extracted the ids for the Tweets, which were at the end of the links and I fed it into Tweepy. The trick here is, Tweepy does not like it when you start asking it for too much so you’ll need to break it down in batches with the function below:

chunks = (total_count — 1) // 50 + 1

for i in range(chunks):

        batch = ids[i*50:(i+1)*50]

        result = fetch_tw(batch)

This way Tweepy scrapes 50 IDs at a time and it will not block you, and you’ll get all the Tweets you need. Sorry Twitter, I promise this is helpful for the greater good, but isn’t that what they all say?

Now that I have fed the Tweet IDs through Tweepy, I have all my 15,000+ Tweets from October 2020. I then continued to do the same for the end of March 2020 in order to have a set of Tweets to compare them to.

Now, we Preprocess!

With all my Tweets scraped, one of the first rules you’ll learn about scraping is it can be very, very dirty. So we first need to clean the Tweets. We need to get rid of URLs, Mentions, Hashtags, lowercase, remove white space and stop-words, and etc. For this study, I rid myself of Smileys and Emojis which I know can contribute greatly to sentiment but due to a time restriction, I needed to work with some pace. I used a combination of a Python library called Tweet-Preprocessing, Gensim, and Regex below:

def preprocess_tweet(row):

    text = row[‘tweet’]

    text = p.clean(text) 
    #clean to normal text removes hashtags and emojis

    text = re.sub(r’[^\w]’, ‘ ‘, text) #Removes all symbols

    text = text.lower() # lowercases all words

    text = re.sub(r’\d+’, ‘’, text) # Removes numbers

    text = re.sub(‘RT[\s]+’, ‘’, text) # Removing RT

    text = re.sub(‘https?:\/\/\S+’, ‘’, text) # Removing hyperlink

    text = remove_stopwords(text) #removes stopwords

    text = re.sub(r’\W*\b\w{1,2}\b’, ‘’, text) 
    #Removes words between 1 and 2 characters short

return text

df_tweets[‘tweet’] = df_tweets.apply(preprocess_tweet, axis=1)

df_tweets[‘tokens’] = df_tweets[‘tweet’].apply(word_tokenize)

[2] S. Srijith, Efficient Tweet Preprocessing (2020), https://www.kaggle.com/sreejiths0/efficient-tweet-preprocessing

Now that my Tweets are squeaky clean onto the Sentiment Analysis!

For the Sentiment Analysis…

…I used TextBlob. It determined two scores: Polarity (positivity/negativity of Tweet) and Subjectivity (subjectivity/objectivity of a Tweet). For this purpose, I only focused on the Polarity score as I wanted to know how Tweeter users felt about Bitcoin at that given time.

[3] Randerson, Twitter Sentiment Analysis (2020), https://medium.com/better-programming/twitter-sentiment-analysis-15d8892c0082

Now that I had the scores I had to determine what meant what. So I created a new column with the given parameters.

1. If the Polarity score was lower than 0, it was a Negative Sentiment
2. If it was equal to 0, it was a Neutral Sentiment
3. Anything else, it was a Positive Sentiment

And here are the results…

Ah-ha! The polarity score is 0.214 at the end of October when the price shot off vs. the polarity score of 0.0734 at the end of March when the price was significantly lower. The October score was nearly 3x the March score! Now we need to look at the actual prices during those times.

Towards the middle/end of March 2020, the price of Bitcoin dramatically dropped to around $5,500

Come end of October 2020, the price dramatically shot up to around $15,000

I even did some topic modeling with Latent Semantic Analysis (LSA), Non-Negative Matrix Factorization (NMF), pyLDAvis, and Corex to observe what most of the Tweets were talking about within that given timeframe. For March 2020 see below:

There is literally a topic that discusses selling Bitcoin which resonates with Bitcoin’s price dip and everyone panic selling.

Then when the price exponentially increased at the end of October, the topics were talking about free giveaways.

Side-note, if you ever see a mention of Bitcoin being given away, run the other way. They are most probably trying to scam you.

In conclusion…

… the people’s sentiment or how they perceive Bitcoin definitely correlates with the price. However, one thing to keep in mind is that this isn’t the only factor. For example, Paypal announced on October 21, 2020, that they would allow its users to Buy/Sell Bitcoin on its platform which I’m sure made a lot of individuals rush to its doors and start buying.

PayPal Launches New Service Enabling Users to Buy, Hold and Sell Cryptocurrency

Then when the price dipped in March, it could have been a combination of many users being liquidated on the crypto margin trading site, Bitmex, and due to a group of Bitcoin “whales” (individuals with a lot Bitcoin) selling off enough of their Bitcoin to move the market (Yes, that happens).

Here's What Caused Bitcoin's 'Extreme' Price Plunge

Whichever impacts the price of Bitcoin, one thing is certain, the sentiment of the people for any sort of currency whether it be the US Dollar or even Bitcoin matters. Depending on the sentiment it will lead to trust or lack there of. So if you are to invest, invest cautiously but always. As Warren Buffet would say…

Lastly…

…please don’t forget to check out my app. It’s a little different from my study because instead of scraping Tweets, I used the NewsAPI to gather articles from the previous month and determine the sentiment based on those. This change in direction was most efficient for the app as scraping 30,000+ Tweets would take a very long time and probably lose the users’ attention. Check it out below!

Please keep in mind none of this information is official financial advice within the blog post or app itself. Thank you!

Please see my other work…

… on my website, Github repository, and LinkedIn to see all my projects. Let me know what I should work on next!

RidwanAlam.com

• ridwan102 - Overview
• Ridwan Alam - Data Scientist - Metis | LinkedIn

How Twitter drives Bitcoin’s price was originally published in Analytics Vidhya on Medium, where people are continuing the conversation by highlighting and responding to this story.

With autonomous vehicles taking off in the past several years, I wanted to explore one aspect that was necessary to create a good autonomous vehicle system, object detection. So what is it exactly? Best way to describe it is the “eyes of car”. As you see below, the vehicles’ cameras are feeding the autonomous system what objects it is seeing. In this case it is a person, pets, and other vehicles. The autonomous system then uses the information to make a decision to turn left, go straight, and etc. Let’s see if we can replicate this Object Detector using Darknet and YOLOv4 to detect traffic signs, traffic lights, and other vehicles.

Process

First I wanted to understand how to approach this project. I was referred to an open source Google Colab notebook by the AI Guy and used that to help springboard my modeling. To create the model I would be using the Darknet open-source neural network and YOLO object detector. To explain the power of YOLO, lets first compare it to a traditional object tracker which utilizes a sliding window.

Traditional Object Detection

As you can observe, the window goes over every part of the image until it detects the actual object, in this case the car. There are two boxes: a Ground Truth Box which was was placed in the picture manually prior to modeling and the other is the Predicted Box which is where the model predicts where the object is. These two boxes are used to calculate the Intersection of Union (IoU) which calculates the Mean Average Precision (mAP) and something we’ll touch upon later in the article. Overall, this process is very computer intensive and inefficient for object detection. In terms of autonomous vehicles, you would not want your car to not recognize a “stop sign” 15 seconds later and then all of a sudden stop. It needs to be instantaneous.

YOLO

Now introducing You Only Look Once or YOLO for short. Joseph Redmon is a computer wiz that created YOLO back in 2015, and he also maintains the Darknet neural network (check out his TED Talk here). What happens in YOLO differently than the above are these three steps: Grid segmentation, Classification, and Image Localization. In this method the model needs to go over an image/video frame one time. Grid segmentation breaks down the picture into evenly sized grid-blocks so every part of the picture is accounted for. Then the model will identify the different classes of the image, in this case dog, bike, and truck. Finally, the objects are located using bounding boxes, as mentioned above, which locate where the objects are within the image, hence the name Image Localization. Putting all that together, you have your model that has successfully identified a dog, bike, and truck and the locations of all of them within an image.

Image Collection and Pre-trained Weights

Before we start building our model, we need to collect some images. I collected all my images of cars, traffic lights, traffic signs, and stop signs from Google Open Images. Before creating the model, I downloaded YOLOv4 Pre-trained weights which are trained on Microsoft’s COCO Dataset of 80 classes. In those 80 classes you have cars, traffic lights, and stop signs. So it would be wise to take advantage of it to increase the mAP (Mean Accuracy Precision). Unfortunately, there are no pre-trained weights for traffic signs hence why it is predicted that the mAP score for it will be lower than the others.

Model Creation and Testing

Now that the model is created, I ran my videos through it and VOILA!

As you can see it picks up all the traffic lights, cars, and most importantly traffic signs! Good, first iteration of the model!

Results: Mean Accuracy Precision

If you recall from above, I mentioned Intersection of Union (IoU) and how that would impact our Mean Accuracy Precision (mAP). To breakdown the IoU once again, please observe the picture of the nice kitten below.

As you can observe, there is a Ground-Truth Bounding Box and Predicted Box. The Ground-Truth Bounding Box is drawn manually before the model is built to indicate exactly where the object is within the picture. The Predicted Box is the model determining where it “thinks” the object is. The greater the intersection between the two bounding boxes the greater the Average Precision (AP) score. An AP is calculated for every single object class in each image and then all the scores are averaged to determine the mAP score, which ultimately decides how well your model is doing.

Please see below for a more visual representation. As you can see the yellow box the “Predicted Box” is 90% overlapping with the Ground-Truth Bound Box.

Finally, see below for individual class mAPs and overall mAP for the model.

• Cars: 80.70%
• Stop Signs: 98.20%
• Traffic Lights: 75.06%
• Traffic Signs: 42.49%
• Overall mAP: 74.11%

As expected the Traffic Signs have the lowest mAP since it did not have any pre-trained weights to train on. I would like to improve this score for future iterations because no one wants to get into an Autonomous Vehicle where it can only predict every 1 out of 2 times if it is a traffic sign or not. However, an overall mAP of 74.11%, not bad for now.

Future Work

I would ideally like to deploy this model on mobile applications as a dashcam app. There is an Android folder I deployed from this open source repo using Android Studio. It was deployed with the YOLOv4 pre-trained weights and as seen below, it can detect cars, trucks, and traffic lights in real time. I customized the app name to “DashKam” and changed the layout, besides that all the code remained the same. Ideally, I would like to deploy this mobile app with my own YOLOv4 custom trained model in a TensorFlow Lite format.

Ethical Dilemma

Joseph Redmon created YOLO to push the boundaries of object detection. With his model we are able to detect objects instantaneously. However, he was approached by a military personnel that informed him they use his model to track vehicles and people, which ultimately can lead to drone strikes. Redmon was so horrified hearing this, that at the beginning of 2020, he proclaimed he would not be doing anymore work with computer vision (which object detection falls under) going forward. What he created was so beautiful and magnificent but it shows that if it falls into the wrong hands it can lead to devastating consequences. For that, I strongly believe we need to hold onto the notion that we in the Data Science community have a responsibility to be ethical in our work for the sake of society.

Other Work

Please visit my website, Github repository, and LinkedIn to see all my projects. Let me know what I should work on next!

RidwanAlam.com

• ridwan102 - Overview
• Ridwan Alam - Data Scientist - Metis | LinkedIn

Special Thanks

• the AI Guy
• hunglc007

Autonomous Vehicle Object Detector with YOLOv4 and Darknet was originally published in Analytics Vidhya on Medium, where people are continuing the conversation by highlighting and responding to this story.

Alright! Back with another scoop within the Data Science World and HOW-IT- CAN-HELP-YOU!

Basically, we’re all tired of delayed flights (pre-COVID times, work with me here folks) and we all wonder 🤔

“Why does American Airlines keep telling me I’m sorry we can’t do anything”

That might’ve just cost me a shot with American

Basically, I wanted to focus on the flight departure delays from the New York City metro area. Why you ask?! Because well, if you’ve ever flown out of there you know how awful it can be. Think about it 20 million people live in the NYC metro area, and they use the 3 airports. Did I mention they all fly international? That’s like merging 20 lanes of cars together. Trust me folks it’s been tried and tested and it doesn’t work, ask LA (I’m burning bridges left and right apparently)

So in this experiment I wanted to focus on if I could create a model that could predict flight departure delays out of La Guardia (LGA), John F. Kennedy (JFK), and Newark (EWR). Here we go.

First I had to gather the data 🤔 which led me to the Bureau of Transportation Statistics. I only focused on the summer months since that’s the busiest travel period, and for this model we’ll only be focusing on the datasets from June, July, and August 2019. After I retrieved the data I really had to think about what features I should be looking at. I settled on departure time (duh), day, carrier, origin, destination, and finally distance. I was pretty satisfied with this list and based on domain knowledge I excluded other features such as wheels up, wheels down, taxing, arrival time, and etc. It would give the model a little bit more of a peek than I would have liked

But first we must balance the data! When I initially looked at it it was leaning a little bit towards the On-Time departures 🤦‍♂

I utilized a couple of different techniques but ultimately settled on Random Over Sampler and SMOTE. Random Over Sampler gave me a nearly, perfectly balanced data set so that was a little too good to be true, so! SMOTE it was!

Now, here comes the fun part, modeling!

No, no, not that type of modeling. This type of modeling:

Alright yes, not as exciting but I promise it gets better (I might be crossing my fingers here).

I looked at a plethora of different models but ultimately settled on Random Forest Classifier and Gradient Boosted Classifier. Random Forest is pretty great out of the box but ended up giving me a much higher score than I had anticipated, especially because I hadn’t tuned any of my hyper-parameters.

What are hyper-parameters you ask?

It’s essentially parameters within the model itself you can adjust, such as max-depth, max features, etc. in order to see what will calculate the most optimal score. Anyways, because of my uncertainty with Random Forest’s score, I decided to lean more towards Gradient Boosted Classifier with a SMOTE balancer plus hyper-parameter tuning. I felt I could justify this more based on my iterative process. I additionally did cross validation where I broke the X-set and y-set into 10 equal parts and the model would be run on one part then another and another and you get a mean average of the accuracy score, which! is not the only score we’ll be chasing. We also want to see if we have good recall, precision, and f1 scores.

Recall = I’m positive I didn’t miss any delayed flights.

Precisions = I’m positive all the flights I labeled delayed were actually delayed

F1 = Weighted average of the two

So after balancing, picking my model, cross validation, and hyper-parameter tuning, I got scores of 90% throughout all of them! Hooray! That means 90% of the time my model will be correct (theoretically)

I’m sure some of you are rolling your eyes and wondering, “Yeah right” and you very well may be correct. I potentially could’ve missed something but that’s the beauty in this process. The Data Science iterative process isn’t just a process to be implemented within the Data Science realm but throughout your everyday process as well, such as your workflow. The point is to get better every time. For myself, the first week I was so lost amongst my peers but now I can feel my legs underneath me and starting to understand concepts better. Throughout the journey thus far, I’ve asked myself

“What can I improve next week”

“How can I learn more effectively”

“How can be more efficient with my time”

“How can I take care of my mental health in order to be more productive as possible for the next week”

This iterative process is unique to you and you alone. It doesn’t matter what the person next to you is doing as long as you know you are getting better on your own path.

I’m sure a lot of you are now wondering “Did I come for a motivational speech?” Well yes, I would like to think I like to motivate people and I hope I can move my peers to be better along with myself.

The reason I bring this up is because my peers at Metis and I all feel similar where we work hard but we question, are we enough? And the answer:

Yes! We are! We are more than enough!

The point of this bootcamp and this journey into Data Science is to learn, create, and make a positive impact on the world through Data Science. That’s why we’ve chosen to do this. So whatever we learn in this Data Science process, we should apply to everyday elements as well.

Alright, back on topic. So the second week, if you read my article, my scores were not great and this week I really wanted to focus on understanding the different models (KNearestNeighbors, Logistic Regression, Gaussian Naive Bayes, Decision Tree Classifier, Random Forest Classifier, and Gradient Boosted Classifier) and I wanted to understand the different ways to balance them (Random OverSampler, SMOTE, ADASYN, and Random Under Sampler). That’s what I set out to do and that’s what I accomplished. No, I do not understand them enough to write a thesis, but I understand them enough to speak to them.

Anyways kiddies, this has been fun but gotta jet! Stay tuned for my next chapter in my Data Ski-Ance journey! Payce!

Departure Delays! Ugh! was originally published in Analytics Vidhya on Medium, where people are continuing the conversation by highlighting and responding to this story.

If you stumbled upon this article thinking we were going trade Pokemon cards and see who could name all 151 of the original Pokemon, you’re about 20 years late. However! If you stumbled upon this page thinking that I was going to share the beginnings of my Data Science journey then you are in the RIGHT place!

I have finally started my Data Ski-Ance journey!! I have been coding for a year and really needed a push. I needed to know that I was good enough and could do this for a living. Welp, I passed two entrance exams for Data Science Bootcamps and now I’m in a pretty good one with Metis. We have lecture, paired programming, projects, challenges, career services, and so on and so forth. The support system has been pretty good so far and even with the $17k price tag it’s really growing on me. So here we go!

We just completed our first project. It was quite difficult especially because it seemed my teammates were more knowledgeable in using Python for Data Science. At that instant I told myself that even if they knew more that it was okay because now I know what I need to improve on. I explained this to one of the TAs and they said “I would recommend you lean into what you don’t know” You know who else said that? Jon Stewart to Trevor Noah before he took over the Daily Show. Jon Stewart advised Trevor Noah to “lean into his fear”. Basically what you should take away from this is that that TA (Thou Who Shall Not Be Named) is a legend.

Anyways, more about Project 1! WomenTechWomenYES!! We were tasked with optimizing the placement of this fictional organization’s street teams in the different New York City subway stations. The street teams would acquire emails to send free tickets for their gala, and at the gala participants would hopefully become donors or be persuaded to pursue a career in tech. The first question was, where do we get the data? To Google! Da-da da-da da-da da-da da-da da-da da-da da-da Goooogleeeee

We searched MTA Turnstile Data and guess what?! It’s all free! Hmmm, could definitely be bad for the MTA considering their financial woes. Wouldn’t want someone to do some sort of a study to expose their shortcomings and post it on the internet, but that couldn’t happen right? Alright, I’m getting ahead of myself. Anyways!

Once we downloaded the data we were tasked with cleaning it. Let’s call it a data “shower” where we get rid of all the junk and gunk and get down to what we really need. This is the part where my teammates took over and I watched because they were moving so fast I didn’t understand what was happening. So I did the second best thing. I took this time to learn from them and jump in when I felt comfortable enough to contribute. Since then I have been able to work on some unfinished problems in Challenge 1 as I started understanding more and more. It was a great feeling!

Continuing on, once my teammates were done cleaning up the data it was time to figure out what to present! Now it’s my time! Presentations baby! One of my teammates had started the presentation but then I took over from there. Having presented data-driven solutions to clients over and over again, this is the part I really enjoyed.

Our strategy: Locate the highest trafficked subway stations and then trim down to target areas for potential donors and students who would consider transitioning to tech as a career.

As you can see from the graph below we had a range of stations to choose from. OOOHHH COLORS

We initially trimmed down to 40 stations to ensure we had a manageable sample size. We were not going to look at all 400+ NYC Subway Stations as that is quite inefficient. From there we managed to locate the donor group in Midtown Manhattan and found correlating data. The mean income in that area is $190,000+…let that sink in for a second. This was definitely going to be one of our top target areas. Also, I want to point out another team used IRS data that has a stat on the most charitable donations for an area. So for a future iteration we could use that as a parameter.

Alright onto the schools! Students are the future and we wanted to ensure they were well incorporated within this study. After looking at some of the busiest stations we noticed how NYU, City College, Borough of Manhattan Community College, and specialized high schools were nearby, BINGO! Got our students!

When it came down to it, we would get the donations we required to support the WomenTechWomenYes organization and we would have the potential students to help increase the participation level of women in tech, BOOM! I think we have ourselves a solution!

Obviously this is by no means perfect but it’s our MVP! No, no not Most Valuable Player, Minimum Viable Product!

Overall, from this experience I learned to work with a team again and really reaffirmed my strengths but realized I needed to double down on my weaknesses. It’s okay because a year ago if you asked me to write you a Python function that calculated (1+1) I would look at you like the way Trump looked when he found out he was going to be the next President of the United States. Dumbfounded.

Anyways, stay tuned for my next issue of my Tech and Data Science journey. Look forward to it!!!