<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Seyed Mousavi on Medium]]></title>
        <description><![CDATA[Stories by Seyed Mousavi on Medium]]></description>
        <link>https://medium.com/@srmousavi25?source=rss-1243c71a2fdb------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*vK3zgBRiQqHQaSz2U_ISgA.jpeg</url>
            <title>Stories by Seyed Mousavi on Medium</title>
            <link>https://medium.com/@srmousavi25?source=rss-1243c71a2fdb------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 11:32:04 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@srmousavi25/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[A Fail-Proof Method for Forecasting Time Series with LSTM]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://blog.gopenai.com/a-fail-proof-method-for-forecasting-time-series-with-lstm-2470d18a825c?source=rss-1243c71a2fdb------2"><img src="https://cdn-images-1.medium.com/max/1018/1*hkM6YPS2FmLLYgww-adRJQ.png" width="1018"></a></p><p class="medium-feed-snippet">The Recap</p><p class="medium-feed-link"><a href="https://blog.gopenai.com/a-fail-proof-method-for-forecasting-time-series-with-lstm-2470d18a825c?source=rss-1243c71a2fdb------2">Continue reading on GoPenAI »</a></p></div>]]></description>
            <link>https://blog.gopenai.com/a-fail-proof-method-for-forecasting-time-series-with-lstm-2470d18a825c?source=rss-1243c71a2fdb------2</link>
            <guid isPermaLink="false">https://medium.com/p/2470d18a825c</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Seyed Mousavi]]></dc:creator>
            <pubDate>Thu, 15 Aug 2024 17:03:17 GMT</pubDate>
            <atom:updated>2024-08-16T10:10:00.266Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Temporal Data in LSTM Time Series Forecasting]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://blog.gopenai.com/temporal-data-in-lstm-time-series-forecasting-c6c61633f915?source=rss-1243c71a2fdb------2"><img src="https://cdn-images-1.medium.com/max/1254/1*FOpbL2PzlVIFiNQ5CLzZbw.jpeg" width="1254"></a></p><p class="medium-feed-snippet">If you noticed in our previous articles, when building LSTM models, we either specified the return_sequences argument to be False or left&#x2026;</p><p class="medium-feed-link"><a href="https://blog.gopenai.com/temporal-data-in-lstm-time-series-forecasting-c6c61633f915?source=rss-1243c71a2fdb------2">Continue reading on GoPenAI »</a></p></div>]]></description>
            <link>https://blog.gopenai.com/temporal-data-in-lstm-time-series-forecasting-c6c61633f915?source=rss-1243c71a2fdb------2</link>
            <guid isPermaLink="false">https://medium.com/p/c6c61633f915</guid>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[time-series-forecasting]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[lstm]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Seyed Mousavi]]></dc:creator>
            <pubDate>Fri, 07 Jun 2024 23:45:53 GMT</pubDate>
            <atom:updated>2024-08-15T17:40:24.041Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[How to perform Grid Search Hyperparameter Tuning for LSTM]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://blog.gopenai.com/how-to-perform-grid-search-hyperparameter-tuning-for-lstm-9bed04932d95?source=rss-1243c71a2fdb------2"><img src="https://cdn-images-1.medium.com/max/640/1*MkbQ0n4mMVgIWkoJdPYsUQ.png" width="640"></a></p><p class="medium-feed-snippet">And also how to use its Early Stop feature</p><p class="medium-feed-link"><a href="https://blog.gopenai.com/how-to-perform-grid-search-hyperparameter-tuning-for-lstm-9bed04932d95?source=rss-1243c71a2fdb------2">Continue reading on GoPenAI »</a></p></div>]]></description>
            <link>https://blog.gopenai.com/how-to-perform-grid-search-hyperparameter-tuning-for-lstm-9bed04932d95?source=rss-1243c71a2fdb------2</link>
            <guid isPermaLink="false">https://medium.com/p/9bed04932d95</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[time-series-forecasting]]></category>
            <category><![CDATA[lstm]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[deep-learning]]></category>
            <dc:creator><![CDATA[Seyed Mousavi]]></dc:creator>
            <pubDate>Thu, 16 May 2024 16:03:18 GMT</pubDate>
            <atom:updated>2024-08-15T17:39:28.209Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[How to fix a common mistake in LSTM time series forecasting]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@srmousavi25/how-to-fix-a-common-mistake-in-lstm-time-series-forecasting-4d4d51d9948f?source=rss-1243c71a2fdb------2"><img src="https://cdn-images-1.medium.com/max/606/1*_hjRp-ffH63A6oMNe7Zg5Q.png" width="606"></a></p><p class="medium-feed-link"><a href="https://medium.com/@srmousavi25/how-to-fix-a-common-mistake-in-lstm-time-series-forecasting-4d4d51d9948f?source=rss-1243c71a2fdb------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@srmousavi25/how-to-fix-a-common-mistake-in-lstm-time-series-forecasting-4d4d51d9948f?source=rss-1243c71a2fdb------2</link>
            <guid isPermaLink="false">https://medium.com/p/4d4d51d9948f</guid>
            <category><![CDATA[lstm]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[time-series-forecasting]]></category>
            <category><![CDATA[deep-learning]]></category>
            <dc:creator><![CDATA[Seyed Mousavi]]></dc:creator>
            <pubDate>Fri, 03 May 2024 02:23:14 GMT</pubDate>
            <atom:updated>2024-08-26T10:23:39.242Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[How to train your LSTM for time series forecasting]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@srmousavi25/how-to-train-your-lstm-for-time-series-forecasting-db724b2aabe2?source=rss-1243c71a2fdb------2"><img src="https://cdn-images-1.medium.com/max/2386/1*RBTqEPabiIVPAqpWlCSKyg.png" width="2386"></a></p><p class="medium-feed-snippet">(and why some of the methods out there are fundamentally flawed)</p><p class="medium-feed-link"><a href="https://medium.com/@srmousavi25/how-to-train-your-lstm-for-time-series-forecasting-db724b2aabe2?source=rss-1243c71a2fdb------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@srmousavi25/how-to-train-your-lstm-for-time-series-forecasting-db724b2aabe2?source=rss-1243c71a2fdb------2</link>
            <guid isPermaLink="false">https://medium.com/p/db724b2aabe2</guid>
            <category><![CDATA[lstm]]></category>
            <category><![CDATA[time-series-forecasting]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[dissertation]]></category>
            <dc:creator><![CDATA[Seyed Mousavi]]></dc:creator>
            <pubDate>Mon, 29 Apr 2024 16:12:20 GMT</pubDate>
            <atom:updated>2024-08-15T17:44:12.319Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Data Science in Aviation]]></title>
            <link>https://medium.com/@srmousavi25/data-science-in-aviation-cc5987a06c04?source=rss-1243c71a2fdb------2</link>
            <guid isPermaLink="false">https://medium.com/p/cc5987a06c04</guid>
            <category><![CDATA[aviation]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[career-change]]></category>
            <category><![CDATA[masters-degree]]></category>
            <category><![CDATA[dissertation]]></category>
            <dc:creator><![CDATA[Seyed Mousavi]]></dc:creator>
            <pubDate>Sat, 09 Mar 2024 14:15:17 GMT</pubDate>
            <atom:updated>2024-03-09T14:15:17.971Z</atom:updated>
            <content:encoded><![CDATA[<p>I have always been fascinated with airplanes. Despite all the physical formulas and all natural imperatives we’ve learned at school, a flying body of 500 tons suspended in the same air we breathe, is perhaps closer to magic than science. Yet it is science, it’s in fact a multidisciplinary net of meticulously studied scientific areas of economics, physics, maths, and business management.</p><figure><img alt="Ahwaz Airport" src="https://cdn-images-1.medium.com/max/1024/1*Mi2H2BBiAlkb2FC-j-Kqxg.jpeg" /></figure><p>One of the modules I passed in my data science course in the university was data visualisation, and the assignment for that module was to formulate a question, find relevant data, explore them and try to answer the question as adequately and as visually as you can. Interestingly, this was just before the time we were supposed to prepare a dissertation proposal. That meant that I had to formulate two research questions, one in the assignment level, and the other is the grave level of a master’s dissertation. All we had done up until then were carefully curated courseworks with clearly defined data problems. But now we had to come up with a research question that’s unique, master level complex but not to the extent of not being able to do it in the timeframe we had, while at the same time answerable with data that can be acquired easily enough.</p><p>I sat with my wife brainstorming whatever ideas came to our minds. Her advantage is that she isn’t a data scientist, sometimes outsiders see things better. Then she suggested something in which airplanes were mentioned and it struck me at once. I don’t even remember her suggestion, I just remember airplane being mentioned. “Based on my experience the best works are those in which you marry what you’re really passionate about with what you’re good at” said Philip our course leader. And that was it. Something I was deeply passionate about, and I had experience in, married with the data skills I’d gained.Of course I saved this for my dissertation and came up with something else for data visualisation.</p><p>After investigating numerous possibilities of applying data science into the aviation industry, I finally settled on what is known as fleet planning. In it’s simplest possible form, fleet planning is the study of finding out which types of aircraft should an air carrier operate. Of course this problem has been studied cover the years and is kinda exhausted. Therefore, this problem is usually coupled with multiple other problems, like fleet assignment (which aircraft should be assigned to which route), carrier’s network, leasing option, hub location problem (HLP), retrofit options, etc.</p><p>While my model integrates all the above mentioned problems, derived from my industry knowledge, it introduces two main novel contributions never explored before. One is carrier type segmentation, where carrier types like Low-cost carriers, Full-service carriers, Cargo carriers, Air taxis and Small certified carriers are recognized and they’re markets are segmented to create a more realistic market forecasting. My model’s second most significant contribution is reconfiguration of aircraft seats. Again never explored before, which is surprising as it is a very practical and convenient business decision.</p><p>I will explain my dissertation in a more technical terms in my next story. So stay tuned!</p><p>All the best.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cc5987a06c04" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Aviation to Data Science, career change in the UK]]></title>
            <link>https://medium.com/@srmousavi25/from-aviation-to-data-science-career-change-in-the-uk-ae0c86c76883?source=rss-1243c71a2fdb------2</link>
            <guid isPermaLink="false">https://medium.com/p/ae0c86c76883</guid>
            <category><![CDATA[student-life]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[career-change]]></category>
            <category><![CDATA[university]]></category>
            <category><![CDATA[london]]></category>
            <dc:creator><![CDATA[Seyed Mousavi]]></dc:creator>
            <pubDate>Tue, 05 Mar 2024 01:50:29 GMT</pubDate>
            <atom:updated>2024-03-05T01:50:29.013Z</atom:updated>
            <content:encoded><![CDATA[<p>For career choices, I’ve worn many hats — from guiding airplanes on the tarmac to tackling fires as a firefighter to managing and supervising construction sites, to designing production lines. But in 2021, I found myself drawn to a new frontier that resonated with my passion for problem-solving, critical thinking, and attention to detail — data science.</p><p>This bold decision led me to embark on a career change, opting to study abroad for a fresh start. After evaluating opportunities in Australia, the Emirates, Germany, and the UK, I chose the latter for its renowned universities and quicker admission process. Little did I know that my expectations for a swift transition from academia to a high-paying job with Visa sponsorship were far from reality.</p><p>After easily acing my IELTS exam and securing admission to two UK universities — the University of Westminster and Middlesex — I chose Westminster without hesitation and soon, I found myself in the bustling city of London.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2V04ypEmR9TAP12sExhdBg.jpeg" /></figure><p>I remember the uneasy apprehension I had before starting my studies, thinking every other person in class surely has some background in data science or at least software engineering except me. Of course that wasn’t the case, but I thought it was. And this misconception made me study even harder, failure is not for me. Never.</p><figure><img alt="“Man is not made for defeat. A man can be destroyed but not defeated.”" src="https://cdn-images-1.medium.com/max/1024/1*XXESGj2eCMKe-FHphb229Q.png" /></figure><p>However, my priority wasn’t just grades; it was providing for my newly formed family. The first two months without work were perhaps the toughest, as I searched everywhere incessently. I remember one terrible day after multiple unsuccessful work applications online, I went outside dropping by whatever business I saw asking if they were hiring. When you are desperate to find something, it’s more fulfilling to at least do it by foot. Your body will know that you are doing your best.</p><p>Those days are far gone now. Fast forward a year later (I wouldn’t bore you with the details of my sweats and toils) I got a good part time job, excelled in my studies, and graduated with distinction.</p><p>Now, I have been searching for work with zero success. the farthest I have gone is two cases where I proceeded into the assessment stages. Nothing else. It’s depressing. And it’s even more depressing considering that I even applied for internships, level 3 roles, apprenticeships, and that I had long forgone Visa sponsorship requirement and decided to apply for a PSW visa myself.</p><p>To combat the disappointments and document my journey as a career-changing data scientist in the UK, I’ve decided to start this weblog. My aim is to share my experiences, connect with those facing similar challenges, and hopefully gather valuable advice from fellow data scientists.</p><p>Stay tuned for more updates!</p><p>All the best.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ae0c86c76883" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>