<?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 Ehsaanalinagyal on Medium]]></title>
        <description><![CDATA[Stories by Ehsaanalinagyal on Medium]]></description>
        <link>https://medium.com/@ehsaanalinagyal?source=rss-bd47b5a53b8e------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*8C8UWNt_Sat8m3xm</url>
            <title>Stories by Ehsaanalinagyal on Medium</title>
            <link>https://medium.com/@ehsaanalinagyal?source=rss-bd47b5a53b8e------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 22:25:34 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@ehsaanalinagyal/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[Job, Apprenticeship, and Internship Websites in France]]></title>
            <link>https://medium.com/@ehsaanalinagyal/job-apprenticeship-and-internship-websites-in-france-618a8e84d6b9?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/618a8e84d6b9</guid>
            <category><![CDATA[jobs]]></category>
            <category><![CDATA[france]]></category>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Thu, 14 May 2026 05:25:55 GMT</pubDate>
            <atom:updated>2026-11-11T13:01:03.405Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://www.hellowork.com/fr-fr/">Hellowork | Offres Emploi, Alerte mail, CV et Recrutement en France</a></p><p>Job teaser</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=618a8e84d6b9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Tidal renewable energy Companies]]></title>
            <link>https://medium.com/@ehsaanalinagyal/tidal-renewable-energy-companies-2ad55cfbda7a?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/2ad55cfbda7a</guid>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Thu, 05 Mar 2026 19:51:58 GMT</pubDate>
            <atom:updated>2026-03-05T19:51:58.855Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://www.linkedin.com/company/openhydro/about/">https://www.greenjobs.ie/companies/open-hydro</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2ad55cfbda7a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Essential Guide for International Students: Settling in Nantes, France]]></title>
            <link>https://medium.com/@ehsaanalinagyal/essential-guide-for-international-students-settling-in-nantes-france-2103cabc8a58?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/2103cabc8a58</guid>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Mon, 22 Sep 2025 17:16:32 GMT</pubDate>
            <atom:updated>2025-09-22T17:16:32.614Z</atom:updated>
            <content:encoded><![CDATA[<h3>Procedures to Follow After Arriving in France (Nantes)</h3><h3>1) Validate Your Visa</h3><p>You must validate your VLS‑TS (long-stay student visa) within 3 months of arrival, through the official website: <a href="https://administration-etrangers-en-france.interieur.gouv.fr./">https://administration-etrangers-en-france.interieur.gouv.fr.</a></p><p>Select ‘Je valide mon VLS‑TS’, enter your visa number (students: CESEDA R431‑16 13°), and complete the payment via electronic stamp (timbre fiscal).</p><p>Notes:<br> — Student visa validation tax is €50 (2025).<br> — For students arriving with spouses (who are not students), the spouse’s visa validation tax is €225.<br> — If you fail to validate within 3 months, your visa becomes invalid — re-entry to France will require a new visa.<br> — If your visa is only valid for 3 months (short-stay), you need to apply for a residence card immediately; the validation procedure does not apply.</p><h3>2) Buy a SIM Card</h3><p>Options include:</p><p>- Major operators: Orange, SFR, Bouygues, Free Mobile.<br> — Budget providers: Lyca Mobile, Lebara, Prixtel, RED by SFR (often require less documentation).<br> — Plans typically cost €10–20/month for 80–100 GB.<br> — Some providers (e.g., Lyca, Lebara) may accept just your passport to activate the SIM.<br> <br> Note: Unless you’re with Orange or SFR, data use in Switzerland and Monaco may be charged heavily — even if you have a ‘European’ package.</p><p>Some sims are already active however not registered so do it via their websites within 5 days</p><h3>4) Health Care (Social Security &amp; Mutuelle)</h3><p>- Register at <a href="https://etudiant-etranger.ameli.fr/">https://etudiant-etranger.ameli.fr</a> for social security coverage.</p><p>- Documents required include Residence proof, Passport, Birth Certificate + FRC, BANK RIB, VISA + Entry Stamp page, Validation of visa Attestation.</p><p>- You’ll receive a temporary social security number, then a permanent one.<br> — With a permanent number, register on <a href="https://www.ameli.fr/">https://www.ameli.fr</a> to request:<br> — Carte Vitale (for reimbursement in France)<br> — EHIC (free for emergencies in Europe)<br> <br> Complementary health insurance (mutuelle):<br> — Social security covers ~70% of medical costs; a mutuelle covers the rest.<br> — Masters students may qualify for free CPAM complementary coverage.<br> — PhD students or working students should consider providers like LMDE or Smerra.</p><h3>3) CAF (Housing Allowance — ALS)</h3><p>- Apply on <a href="https://caf.fr./">https://caf.fr.</a><br> — Required documents: passport, visa validation, proof of accommodation, RIB, CAF attestation or rental contract.</p><p>-You will create a temporary login and after signin, go to housing assistance. ‘Faire le Demande’</p><p>- For IMT Students a proper guide with pictures is separately available. Other students can use similar steps, however do change it according to your university.</p><p>- CAF payments usually go directly to CROUS or your landlord and can take ~2–3 months to process. Allowance is backdated to the month after application. Once processed again go to declare resources for last year as ZERO to display Family Quotient. <br> — Always verify that your residence is CAF-eligible before renting.</p><h3>5) Opening a Bank Account</h3><p>Recommended banks include:<br> — Société Générale<br> — BNP Paribas<br> — LCL<br> Steps: Take an appointment (Rendez-vous) online or by visiting branch. Before visiting use help from any seniors to send you a link for referral (Parrianage) for respective bank of your choice. Fill details using the link to open account however it will ask/direct you to take appointment physically as you do not have French ‘titre de Sejour’. The person sending you link, and you will get around 50 or 80 euros this way. Also, the seniors can guide you for app and usage later. Just ask someone in group for respective bank of your choice to send you referral link code.</p><p>On Appointment day: Required documents are as follows:<br> — Passport<br> — Proof of residence (rental contract or attestation de residence or Quittance de loyer )<br> — Proof of enrolment in university (registration certificate)</p><h3>6) Tram/Bus Card (Transport)</h3><p>- Nantes offers an extensive tram and bus network via TAN (Transports de l’Agglomération Nantaise).</p><p>- Eservices nantes platform can be used for online procedure to apply for card. Link: <a href="https://eservices.nantesmetropole.fr/accueil-particuliers">https://eservices.nantesmetropole.fr/accueil-particuliers</a><br> — Monthly passes for students (under 26) typically cost €20–25.<br> — One-time card fee is around €5.<br> — Tickets can be bought at vending machines, stops, or via the Neolib app.<br> — Traveling without a valid ticket may result in a fine of €60–110.</p><p>Free/subsidized transport can be applied after CAF file is processed and you receiv Family Quotient. However, for students below 25 years Tax declaration documents are required. You must visit Central Town Office to apply for it afterwards. Documents required are Passport, Attestation of payment from CAF indicating Family Quotient, Attestation of residence, Picture, RIB.</p><h3>7) Tax Declaration</h3><p>- The tax portal opens annually in April–May on <a href="https://www.impots.gouv.fr./">https://www.impots.gouv.fr.</a><br> — First-time filers should submit Formulaire 2042 on paper (tick the first declaration box).<br> — Required info: salary (box 1AJ), withholding tax (box 8HV).<br> — Attach passport, visa validation, proof of address, RIB, and social security certificate, then drop at the tax office or mail through La Poste (free) or better to visit physically before 12:00. It is recommended to visit Tax office in person and ask them to open tax rights (Overture de droit). This can be done at any time of year with same documents mentioned above.</p><h3>8) CROUS Housing Renewal</h3><p>- The renewal window runs from mid‑April to mid‑May on <a href="https://www.messervices.etudiant.gouv.fr./">https://www.messervices.etudiant.gouv.fr.</a><br> — Requires an updated VISALE certificate.<br> — Missing the deadline means applying as a new candidate.</p><h3>9) Visa Renewal &amp; Job Search Visa (RECE)</h3><p>- Apply 90–120 days before visa expiry on <a href="https://administration-etrangers-en-france.interieur.gouv.fr./">https://administration-etrangers-en-france.interieur.gouv.fr.</a><br> — Documents: passport, proof of residence, RIB, social security certificate, and approximately €8,000 in bank funds (students), or €16,000 for spouses.<br> — For the Job Search Visa (RECE), follow the same process; bank statement is not required.</p><p>In August prefecture is extremely slow in processing visa files, so consider them in advance.</p><h3>General Information</h3><h3>Insurances</h3><p>- Civil liability (responsabilité civile) and housing insurance are essential — usually €20–40/year.<br> — Options available via etudassur and similar websites.</p><h3>Groceries &amp; Halal Food</h3><p>- Halal meat Chicken (Poulet), Meat (Viande) is available easily in Halal Boucheries across Nantes.</p><p>- Search for GARCHA INDIA (For Asian Masala’s and Desi Punjab AATA — flour) or AFGHAN BUCHERIE (Meat, chicken, Rice and Asian Masala’s)</p><p>Cost: 10 kg Aata (Flour) = 16 to 18 euros, 5kg Rice = 15 euros, 10 kg = 27 euros</p><p>Leclerc , Carrefour, Auchan and LIDL are supermarkets for general groceries. Usually, students prefer LIDL and Leclere.</p><h3>Mosques &amp; Prayer</h3><p>- For IMT and Polytech nantes students closest mosque is MOSQUE AL-MOHSININE Near ‘Pin Sec’ Stop TRAM1.</p><p>- For ECN and students living near Recteur Schmitt area, closest option is MOSQUE ARHAMA near ‘Rennes cassin’ stop TRAM2</p><p>- For remaining there is also GRAND MOSQUE de Nantes near ‘Marie de Doulon’ stop TRAM1.</p><h3>Medical &amp; Doctor Services</h3><p>CHU Nantes can be accessed for emergencies, but it is recommended only in extreme cases. Also, Dental services can be availed at CHU Nantes hospital but are recommended after you have received your social security and mutual.</p><p><strong>IMPORTANT</strong>: Wait for social security number and then declare your medicine generalist as soon as possible — Use <a href="https://www.doctolib.fr/">https://www.doctolib.fr</a> to register with a family doctor (médecin traitant) — filter for English-speaking providers. Also search for ‘Medicine Generaliste’ from google maps and request them to be your medicine generalist. If one agrees, take appointment. And visit him/her for a general checkup and declaration. Do this ASAP after receipt of your social security number and before you are ill.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2103cabc8a58" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ How I Applied to a Public University in France (Master’s) — Step-by-Step Guide Based on My…]]></title>
            <link>https://medium.com/@ehsaanalinagyal/how-i-applied-to-a-public-university-in-france-masters-step-by-step-guide-based-on-my-9c9b7d607b98?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/9c9b7d607b98</guid>
            <category><![CDATA[masters]]></category>
            <category><![CDATA[france]]></category>
            <category><![CDATA[top-public-university]]></category>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Wed, 06 Aug 2025 09:09:12 GMT</pubDate>
            <atom:updated>2025-08-06T09:09:12.329Z</atom:updated>
            <content:encoded><![CDATA[<h3>🇫🇷 How I Applied to a Public University in France (Master’s) — Step-by-Step Guide Based on My Real Experience | Annual Fee: Only €250!</h3><p>Hey everyone!</p><p>I’m currently heading to France for my fully funded PhD, but before that, I also went through the full process of applying to a Master’s program at a public university in France. I received admission to Université de Bourgogne in their Master Recherche Health-AI (M2) program — with an annual fee of just €250.</p><p>Here’s the complete step-by-step guide to applying through Campus France Pakistan, including interview questions, documents, fees, visa, and final steps. Let’s dive in! 🔍</p><p>✅ STEP-BY-STEP APPLICATION PROCESS</p><p>🔹 Step 1: Create a Campus France / Études en France Account</p><p>Go to: <a href="https://pastel.diplomatie.gouv.fr/etudesenfrance">https://pastel.diplomatie.gouv.fr/etudesenfrance</a></p><p>🔹 Step 2: Fill Out the Application Form</p><p>Upload all your personal and educational documents.</p><p>🔹 Step 3: Fill Out the Program Cart</p><p>Search for English-taught programs relevant to your field (e.g., Data Science, Machine Learning, Computer Science, Health AI, etc.)</p><p>You can select a maximum of 7 universities.</p><p>Be careful while choosing — you’ll see programs ranging from €250 to €18,000+ per year.</p><p>If you have a strong academic profile (CGPA, background, etc.), you can target public universities with €250/year fees.</p><p>✔️ In my case, I got accepted to Université de Bourgogne with €250/year tuition.</p><p>🔹 Step 4: Submit File to Campus France Pakistan</p><p>This means Campus France will start verifying your uploaded documents from Step 2.</p><p>🗓️ When to Apply?</p><p>For September Intake, Campus France portal usually opens in October and remains open till 31st December.</p><p>After 31 Dec, you can only apply directly to university websites, which is possible — but applying via Campus France makes your future visa process easier.</p><p>📝 AFTER VERIFICATION BY CAMPUS FRANCE</p><p>✅ Pay Campus France Fee — PKR 30,000 (Mandatory)</p><p>🎤 Campus France Interview</p><p>Bring:</p><p>Passport</p><p>All educational documents (original + photocopy)</p><p>⚠️ Common Interview Questions:</p><p>Introduce yourself</p><p>Why France?</p><p>Why this particular university (e.g., ICN)?</p><p>O-Levels / Matric marks?</p><p>Intermediate marks?</p><p>Why Inter after O-Levels?</p><p>Bachelor’s CGPA?</p><p>Who is your sponsor?</p><p>What does your father do?</p><p>What does your mother do?</p><p>What are your future goals?</p><p>Expected monthly expenses in France?</p><p>After the interview, Campus France will submit your application to the 7 selected universities.</p><p>You will start receiving admission results (acceptance or rejection) from universities via Campus France — typically until the end of May.</p><p>If you apply directly to universities (not via Campus France), results may come later or on their own schedule.</p><p>✅ Once You Get Admission</p><p>Log into Campus France portal and choose the university you want to continue with.</p><p>📅 Book AEG Appointment</p><p>AEG appointment fee: Around PKR 2,500</p><p>You can book from AEG offices in Islamabad, Lahore, or Karachi</p><p>🏛️ Embassy Visit for Visa Application</p><p>After your AEG appointment:</p><p>✈️ Go to the French Embassy with the following documents:</p><p>🗂️ Visa Document Checklist:</p><p>2 Visa Application Forms + 3 Passport-size Photos</p><p>Passport Copy</p><p>CNIC Copy</p><p>2 Updated CVs</p><p>Admission Offer Letter</p><p>EEF Fee Receipt</p><p>Cover Letter</p><p>All Educational Documents</p><p>• English Proficiency Letter or IELTS (if required)</p><p>• Optional: Recommendation Letters &amp; Certificates</p><p>Sponsor Affidavit</p><p>Sponsor’s Bank Statement + Maintenance Letter + Income Proof</p><p>Accommodation Proof</p><p>Flight Ticket (if available)</p><p>Supporting Documents (if any)</p><p>FRC (Family Registration Certificate)</p><p>Birth Certificate</p><p>💰 Visa Application Fee: €50 (confirm updated fee while filling visa form)</p><p>🖊️ Apply online: <a href="https://france-visas.gouv.fr/en/">https://france-visas.gouv.fr/en/</a></p><p>🔎 Embassy may take your biometric data and conduct a short interview (In my case, there was no interview).</p><p>⏳ Visa Decision</p><p>After submission, your visa portal status will update.</p><p>In my case, I got my visa decision in 1 day!</p><p>Once approved, AEG will call you to collect your passport.</p><p>📌 Check if the visa is granted. If rejected, a reason letter will also be attached.</p><p>#StudyInFrance #MastersInFrance #CampusFrance #ÉtudesEnFrance #PublicUniversityFrance #VisaGuide #PhDInFrance #AEGAppointment #FranceVisa #AffordableEducation #PersonalExperience #PakistaniStudentsInFrance</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9c9b7d607b98" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Dementia Disorder Detection using Machine Learning]]></title>
            <link>https://medium.com/@ehsaanalinagyal/dementia-disorder-detection-using-machine-learning-a37cb03f0e9d?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/a37cb03f0e9d</guid>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Fri, 24 Jan 2025 09:28:01 GMT</pubDate>
            <atom:updated>2025-01-24T09:28:01.337Z</atom:updated>
            <content:encoded><![CDATA[<p>Abstract — Dementia is a neurodegenerative disorder of the motor system in the nervous system characterized primarily by the loss of memory and social dysfunction. Despite the fact that it is one of the most widespread mental health conditions in the world, it is still not diagnosed or treated in a timely manner. Statistics have shown that by 2030 more than 82 million people will suffer from dementia. In the past decades, researchers have made substantial progress by developing machine learning (ML) algorithms to detect and diagnose neurodegenerative diseases with the power of computers. We have seen many papers analyzing the clinical implications of machine learning methods i.e. SVM, J48, ANN, Decision Tree, Multilayer Perceptron, and KNN. Studies have shown that the J48, Support Vector Machine and Random Forest outperform other techniques for detecting dementia, with 99.52% and 92.36% accuracy, respectively. While the benefits were modest, many of the statistical feature selection strategies tested certainly performed better than others when it came to rating the features.</p><p>Introduction Dementia is a condition in which the function of brain cells deteriorates over time and eventually leads to death. It is a rare disease that has not yet been properly treated. The disease is a combination of many other diseases such as Alzheimer’s, levy-body dementia, fronto-temporal and other diseases. Statistics have shown that more than 82 million people will be suffering from dementia by 2030. Currently Turkey is the one who has major cases of dementia with the cases ratio of up to 60% and in Pakistan the ratio is about 20% and on 114 number in the world. As in the symptoms of dementia includes forgetfulness, short-term memory loss etc. Generally people take it for granted and relate it with age i.e. after passing 50 you could normally have this issues neglecting the fact that it could actually be dementia. Currently different detection approaches are proposed using sensor, through which [1] detect the behavior of the patients and then applied classifiers on it to detect the variation. But then the issue arises about the training of the model some are training their models on finite number of individuals [9]. Mostly work is done in general dementia detection, but some are working to give prediction about whether the person is having dementia, if yes then what is the severity level from extreme to normal [2]. Different techniques are being used so far some are using questionnaire approach through which they ask a range of questions and based on them the detects the level of severity of the disease. Others are trying to detect the disease using MRI imaging, extract features and then apply classifiers and get the result. In MRI, some have introduced 3D imaging [4] while other are working on the 1D images and getting accurate results. The detection is not only limited to imaging; researchers are also working on home based remote assessment of the patients [6] this assessment is not just made in few hours but some have even collected the information of the patients for about consecutively in 3 years Speech recognition is also getting popularity in this field, but not much work has been done on this. Some researchers have conducted test on finite number of individuals and gotten impressive results. Not only speech, but also detection is carrying out using eyes. Although these are novel approaches. But in this review we are proposing new areas to explore.</p><p>Literature Review A lot of research has been conducted for the detection of dementia in patients to help the caregivers using deep learning. [1] Sensor based deviant behavior detection system was proposed for the caregivers through which they can manage patients remotely. The system utilized an auto-encoder and long shortterm memory (LSTM) model which was the combination of unsupervised learning through which the characteristics of patients were extracted and through LSTM it identified the deviancy from the final output measure with accuracy of 96% and 99% respectively. However, the data collected for this experiment was limited so the model could have predicted wrongly as well. [2] proposed an informant based questionnaire approach for dementia diagnosis, in which they used deep neural networks along with Keras Api and compared their model with other traditional approaches. They divided patients randomly and tested them based on accuracy, precision, recall, and F1-Score, and were able to achieve the best result with accuracy 0.88 along with the same f1-score. Normal, mild, very mild cognitive dementia, moderate dementia, and severe dementia were all effectively diagnosed using the model. It was basically a bounded survey approach having limited questions with taking a lot of time. Another approach was introduced by [3] through camera they detected the skeleton of humans and captured necessary information about it. After detection they applied a convolutional neural network classifier to get important features from these videos and based on that they detected dementia and achieved sensitivity of 74.10%. But the model was unable to detect dementia symptoms accurately and was consuming so much time. [4] proposed a fast and cost-effective approach for the detection of dementia using convolutional Neural networks along with 3D MRI imaging. They proposed this approach comparative to those conventional techniques where feature extraction was manual. The model is trained on Floydhub’s GPU with accuracy of 80.25%. The model is lacking certainty which is something of a big issue in diagnosis. [5] have presented a comparative solution to the questionnaire based technique for dementia diagnosis. They proposed a deep learning solution to this issue by 1D Convolutional Neural Network and LSTM classifier to detect dementia with speech. The training is conducted on 250 epochs and has accuracy 90.8% along with f-measure 89.7%. The model was really costly, even the same accuracy could be achieved using feature vectors which is cost effective.</p><p>Based on behavioral assessment [6] proposed another great approach which is less time consuming and takes 2 hours for thorough dementia detection. The proposed system was basically based on the idea of a smart home where assessment is easier using ambient sensors and the datasets used for this purpose were 2. One contains the data from 232 patients while the other one received data from 30 patients and using an ambient sensor, the model was able to successfully achieve precision and recall ratio of 98% for first and 89.9% for second dataset respectively. Since the model was trained and tested on limited number of people it could have based on a motion sensor and tested on wide range to get better understanding. [7] have analyzed extreme learning machine technique to distinguish between mild, Alzheimer and normal control. They utilized a recursive support vector machine (SVM) approach of feature elimination to extract the features from the subset and get the better result compared to traditional SVM and get the accuracy of about 86 to 92%. [8] presented a model which will work on a 5g model that detects the dementia people from the normal with deep learning techniques including convolutional neural network along with Restricted Boltzmann machines and auto encoder. The model is accurate but the issue is that it totally depends on the 5G which is not available everywhere. [9] used four classifiers Naive Bayes, random forest, Jrip, Rules based and decision tree. They used MRI images and used segmentation to achieve good results. Among all these techniques the best result of 100% accuracy is achieved using Jrip with the dataset of 466 men and women. But this limitation as well as the dataset is local (Indian) and limited so the result couldn’t be said as best. [10] has presented the research on the topic of dementia disorder detection, they opted for the latest speech feature technology for detection along with the speech data gathered from 26 people and achieved accuracy 81% along with f1- score 74%. The only limitation of this work is that they have worked with a limited number of patients which can’t give an accurate result. [11] have presented Fractal Texture Analysis (FTA) technique for the detection of dementia along with the dataset of OASIS with the accuracy of 97.5%. The decomposition was done using a two-threshold binary decomposition technique. A feed forward artificial neural network was used to classify dementia patients along with OASIS dataset. [12] proposed Deep MRS network which was trained using data from 135 patients gathered at the University Hospital of Poitiers. They achieved classification of patients with early Alzheimer’s disease vs healthy controls has an AUC of 94, 74%, a sensitivity of 100%, and a specificity of 89, 47%, indicating a promising early dementia detection performance. [13] have presented the study which demonstrates the value of self-supervised representation learning techniques in medical applications where a small number of patients and the complexity of the setting make applying state-of-the-art methods difficult. They are still working and haven’t mentioned specific figures on it. [14] have used deep learning based model along with keras API, they were focused to differentiate the healthy people from dementia patients and achieved good result with accuracy of 96.3%, sensitivity 94.9% respectively. Since they were working on mild one the it could also have enhanced on different other dementia types as well. [15] For SVM and regularized logistic regression, they used the ADNI-GO/2 study with PET scans. They have not mentioned any specific accuracy but the model performed well compared to conventional approaches. [16] analyzed structural issues and classifies the pathology using information of brain asymmetry collected from different MRI images in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. The trials show that there is a difference between people with non-dementia and patients with mild dementia, they were successful and achieved accuracy of 92.5% using CNN.</p><p>Conclusion</p><p>Nowadays, a lot of work has been done on a mild detection with limited datasets and traditional classifications approaches. There is a need to work on innovative techniques such as earlier robotic detection which is still not yet finalized. And also past researchers have tested on small data sets. Beside that the work is only conducted on the traditional way of the detection of the dementia. There is a need to work on the fields such as detection through eyes or more focused all the sub types of the dementia rather than just focusing on the general dementia with existing classifiers. Should be worked with dynamic techniques.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a37cb03f0e9d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Critical Review on Paper Titled:
“Depression detection from & social
network data using machine…]]></title>
            <link>https://medium.com/@ehsaanalinagyal/critical-review-on-paper-titled-depression-detection-from-social-network-data-using-machine-7ea5d8bc5c90?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/7ea5d8bc5c90</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[depression-detection]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Fri, 24 Jan 2025 07:34:19 GMT</pubDate>
            <atom:updated>2025-01-24T07:37:20.540Z</atom:updated>
            <content:encoded><![CDATA[<h3>Critical Review on Paper Titled:<br>“Depression detection from &amp; social network data using machine learning techniques”</h3><p>Critical Review on Paper Titled: “Depression detection from &amp; social network data using machine learning techniques”<br>Authors — Md. Rafiqul Islam, Muhammad Ashad Kabir, Ashir<br>Ahmed, Abu Raihan M. Kamal, Hua Wang and Anwaar Ulhaq</p><p>Submitted By:<br>Ehsaan Ali, Azka Yamin, Danish Haider, Faizan Hassan Siddiqui, Syed Ahmed Ali, Muhammad Haris Waqar</p><p>Depression is a mental health problem affecting millions worldwide (4.4 percentage population worldwide). It is a major public health concern as progressive depression often leads to suicide. Approximately 80,0000 deaths occur due to suicide every year and it has become the second leading cause of death among youngsters (15–29 years old) . The reviewed paper aims to detect depression using machine learning techniques from Facebook user data. The dataset used has been collected and prepared from Facebook user comments by the authors themselves. <br>The aim of this critical review is to analyze the given research paper thoroughly including the general write up, structure, methodology adopted, results, presentation of results and the subsequent discussions. In this review we will be describing the content of the article as a summary in section II. The discussions and comments on the paper are presented in section III where all the major portions of the paper are discussed individually. Finally, section IV provides a brief conclusion of this review.</p><p>2 Summary of Article Content<br>The advent of internet and social media has changed how people interact with each other. Online social platforms like twitter, Facebook, Instagram etc. have enabled people to share their thoughts, ideas, experiences, and emotions with people across the world.<br>This freedom of speech and outreach has made a global impact by giving people a safe space to vent, communicate and socialize, moreover, it has provided health workers an opportunity to analyze and deduce about the mental health of the users. The purpose of this paper is to detect depression using machine learning (ML) techniques from Facebook user comments data. The paper also aimed to find the factors that can be used in the identification of depression and the suitable ML techniques to apply.<br>The research was focused on the following three types of factors extracted from the comments and posts: temporal process, emotional process, and linguistic style features.<br>The dataset was collected from Facebook pages by using NCapture software. LIWC software was used for extracting features from the raw Facebook data. 21 features were extracted using LIWC from which 13 represent linguistic features such as articles, verbs, conjunctions, personal pronouns, verbs etc., 5 represent Emotional process (sad, anger, anxiety) and 3 represent temporal process (past, present, future). The ground truth labels for the dataset were prepared and out of 7145 comments, 58 percentage contained yes (depression present) and 42 percentage no (depression not present).<br>These 21 attributes from the 3 factors were used to for prediction of depression by KNN, SVM, decision tree and ensemble classifier. All the models were trained on the above mentioned three features independently and jointly (all 3). Accuracy of about 60–80 percentage was achieved by using these classifiers. The decision tree classifier gives the best results as compared to other models considering the precision, recall and F-measure scores. The authors aim to use more features and dataset in future work and stress on the need for more studies on depression analysis.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7ea5d8bc5c90" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ Best Websites & Resources to Find a Fully Funded PhD in France]]></title>
            <link>https://medium.com/@ehsaanalinagyal/good-website-resource-to-find-phd-position-in-france-f84228a3c142?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/f84228a3c142</guid>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Tue, 22 Oct 2024 10:53:18 GMT</pubDate>
            <atom:updated>2025-08-02T18:43:55.039Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>✍️ By Ehsaan Ali Nagyal</strong><br> 📅 <em>Oct 22, 2024</em></p><p>Looking for a <strong>fully funded PhD position in France</strong>? After navigating the process myself and securing admission, I’m sharing the exact <strong>websites and resources</strong> that helped me find funded opportunities across top French institutions.</p><p>Whether you’re applying from Europe, Asia, or anywhere else, these platforms can help you find PhD positions in both academic and industrial sectors.</p><h3>🌐 Top General Platforms</h3><p>Here are the best websites to start your PhD search in France:</p><p>🔗 ABG (Association Bernard Gregory)<br> 🔗 EURAXESS France<br> 🔗 <a href="https://euraxess.ec.europa.eu/">EURAXESS Europe</a><br> 🔗 <a href="https://scholarshipdb.net/">ScholarshipDB — France PhDs</a><br> 🔗 <a href="https://academicpositions.fr/jobs/position/phd/country/france">Academic Positions — France</a><br> 🔗 Thèses en Bretagne Loire (TBL)<br> 🔗 Université Paris-Saclay PhD Portal<br> 🔗 <a href="https://jobs.inria.fr/public/classic/en/offres?filtre=doctorants">INRIA PhD Job Portal</a><br> 🔗 <a href="https://emploi.cnrs.fr/Recherche.aspx?bassin=INDEF&amp;type=ITACDD&amp;texte=">CNRS Job Portal</a><br> 🔗 <a href="https://fr.indeed.com/q-these-cifre-emplois.html?vjk=659dc034c48b0f7e">CIFRE PhD Jobs — Industry Collaboration</a><br> 🔗 <a href="https://www.litislab.fr/emplois">LITIS Lab — PhD Positions</a></p><h3>🏫 Major Universities with PhD Portals</h3><p>Many French universities list PhD opportunities on their own websites. Here are a few reliable ones:</p><ol><li>🎓 Sorbonne University</li><li>🎓 University of Paris (Doctoral Schools)</li><li>🎓 École Normale Supérieure (ENS)</li><li>🎓 École Polytechnique — PhD Programs</li><li>🎓 University of Bordeaux — Doctoral Schools</li></ol><p>Pro tip: Many positions are offered through “Doctoral Schools” or “Écoles Doctorales.” Always check those pages on each university’s website.</p><h3>🎓 Why Choose France for Your PhD?</h3><ul><li><strong>Fully funded PhD contracts</strong> (with monthly salary)</li><li><strong>No or minimal tuition fees</strong> at public universities</li><li><strong>Cutting-edge research labs</strong> (CNRS, INRIA, etc.)</li><li>Many programs are <strong>taught in English</strong></li></ul><h3>💡 Tips for Applying</h3><p>🗓️ Apply between <strong>January–June</strong> for most October start dates<br> 📄 Tailor your <strong>CV and motivation letter</strong> to each position<br> 📧 Reach out to <strong>supervisors or lab heads</strong> if contacts are listed<br> 📊 Keep a simple <strong>spreadsheet</strong> to track deadlines, documents, and responses</p><h3>📌 Final Thoughts</h3><p>France is a great destination for international researchers. These platforms make it much easier to explore the right opportunities — whether in academia, public labs, or industry (via CIFRE). Stay persistent, and good luck with your applications!</p><p>🔁 Share this post if it might help someone in your network.<br> 💬 Questions? Feel free to drop a comment or connect!</p><p>#PhDInFrance #FundedPhD #ResearchOpportunities #StudyInFrance #InternationalStudents #AcademicJobs #Doctorate #EURAXESS #CIFRE #INRIA #CNRS</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f84228a3c142" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[HealthCare Dashboard Design Using Python Django and HightChart API]]></title>
            <link>https://medium.com/@ehsaanalinagyal/healthcare-dashboard-design-using-python-django-and-hightchart-api-9fe5d2fc5063?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/9fe5d2fc5063</guid>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Thu, 03 Oct 2024 08:54:24 GMT</pubDate>
            <atom:updated>2024-10-03T08:54:24.720Z</atom:updated>
            <content:encoded><![CDATA[<p>TB section</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gcc_nKy58nJVARFdk02bfg.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9fe5d2fc5063" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Construction Activity Recognition-using deep learning]]></title>
            <link>https://medium.com/@ehsaanalinagyal/construction-activity-recognition-using-deep-learning-ed4e63aeb6a1?source=rss-bd47b5a53b8e------2</link>
            <guid isPermaLink="false">https://medium.com/p/ed4e63aeb6a1</guid>
            <dc:creator><![CDATA[Ehsaanalinagyal]]></dc:creator>
            <pubDate>Thu, 03 Oct 2024 08:21:38 GMT</pubDate>
            <atom:updated>2024-10-03T09:03:11.541Z</atom:updated>
            <content:encoded><![CDATA[<p>High definition (HD) video cameras have been used to record daily activities at construction jobsites into videos. These videos contain rich workforce and equipment information for site engineers and project managers to analyze construction productivity, monitor construction progress, inspect jobsite safety, etc. It is difficult to extract this information from the videos with higher performance. In this study, a state-of-the-art technique, which combines the CNN and LSTM to extract spatial and temporal information respectively, and perform the activity recognition on construction sites. The proposed model successfully recognized the six activities on construction sites with over 90% of F1-score.</p><p><strong>Output</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/687/1*iXE1-GHWOK-WR4wY8bICgQ.png" /></figure><p>Code link</p><p><a href="https://github.com/Ehsaanali/Construction-Activity-Recognition-using-deep-learning">GitHub - Ehsaanali/Construction-Activity-Recognition-using-deep-learning: High definition (HD) video cameras have been used to record daily activities at construction jobsites into videos. These videos contain rich workforce and equipment information for site engineers and project managers to analyze construction productivity, monitor construction progress, inspect jobsite safety, etc.</a></p><p>Dataset link</p><p><a href="https://github.com/Ehsaanali/Construction-Activity-Recognition-using-deep-learning/tree/main/Custom%20Dataset">https://github.com/Ehsaanali/Construction-Activity-Recognition-using-deep-learning/tree/main/Custom%20Dataset</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ed4e63aeb6a1" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>