<?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 Apshana S P on Medium]]></title>
        <description><![CDATA[Stories by Apshana S P on Medium]]></description>
        <link>https://medium.com/@apshana?source=rss-545dd27682cf------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*sydA_LnPpSG7glOh</url>
            <title>Stories by Apshana S P on Medium</title>
            <link>https://medium.com/@apshana?source=rss-545dd27682cf------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 19:42:37 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@apshana/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[Case Study: Leveraging Web Mining for Enhancing Online Retail Customer Experience]]></title>
            <link>https://medium.com/@apshana/case-study-leveraging-web-mining-for-enhancing-online-retail-customer-experience-78dd4bef62fc?source=rss-545dd27682cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/78dd4bef62fc</guid>
            <dc:creator><![CDATA[Apshana S P]]></dc:creator>
            <pubDate>Sat, 10 May 2025 16:45:10 GMT</pubDate>
            <atom:updated>2025-05-10T16:45:10.312Z</atom:updated>
            <content:encoded><![CDATA[<p>Case Study: Leveraging Web Mining for Enhancing Online Retail Customer Experience</p><p>1. Introduction<br>The exponential growth of e-commerce platforms has led to the accumulation of vast amounts of user-generated data on the web. Web mining, a subset of data mining, allows businesses to extract meaningful patterns from this unstructured data. This case study explores how an online retail company used web mining techniques to improve customer satisfaction and retention through personalized recommendations and behavior analysis.</p><p>2. Background<br>Company X is a mid-sized online retailer specializing in lifestyle products, including apparel, electronics, and home decor. Despite consistent web traffic, the company noticed a decline in repeat purchases and increasing bounce rates on its product pages. The management team decided to implement web mining techniques to gain deeper insights into user behavior and improve overall engagement.</p><p>3. Objectives<br>The primary objectives of the web mining initiative were:</p><p>To understand customer navigation patterns on the website.</p><p>To personalize product recommendations based on user behavior.</p><p>To identify high-exit pages and optimize them for better conversion.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/624/1*7KYHnxdN52l8G2rPSrupjA.png" /></figure><p>4. Methodology<br>The company adopted a three-pronged web mining approach:</p><p>a. Web Content Mining<br>By analyzing product reviews and user comments, the company used sentiment analysis to gauge customer satisfaction and identify common pain points related to product quality, shipping, and user interface.</p><p>b. Web Structure Mining<br>The link structure of the website was analyzed to understand how users transitioned from one product or category page to another. The PageRank algorithm helped identify key product pages that served as central hubs in user navigation.</p><p>c. Web Usage Mining<br>Server logs and clickstream data were processed using clustering algorithms (e.g., K-means) to segment users into behavior-based groups: first-time visitors, frequent browsers, and high-value customers. Association rule mining was employed to uncover product bundling patterns (e.g., customers who viewed a laptop also viewed laptop sleeves).</p><p>5. Implementation and Tools<br>The project utilized open-source tools like Apache Hadoop for distributed data processing, Python for scripting, and Weka for machine learning. Google Analytics data was also integrated for additional behavioral insights.</p><p>6. Results and Insights</p><p>Improved Recommendations: Implementing a collaborative filtering algorithm based on user behavior led to a 22% increase in product recommendation click-through rates.</p><p>Page Optimization: High-exit pages were redesigned based on user heatmaps and A/B testing, resulting in a 15% improvement in conversion.</p><p>Customer Retention: Personalized emails based on browsing history increased repeat purchases by 18% over three months.</p><p>7. Challenges<br>Some challenges included handling noisy data, ensuring user privacy compliance (e.g., GDPR), and the need for real-time analysis, which required infrastructure upgrades.</p><p>8. Conclusion<br>This case illustrates the transformative potential of web mining in enhancing user experience and business performance. By humanizing data through behavior patterns and preferences, web mining enables organizations to respond more intuitively to customer needs, fostering stronger digital relationships.</p><p>9. Future Work<br>Future enhancements include integrating social media mining for more holistic customer profiling and deploying deep learning models for more nuanced sentiment analysis and recommendation systems.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=78dd4bef62fc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[CASE STUDY FOR 4 COMPANIES ON BIG DATA : UBER , EQUIFAX , WALMART,MARRIOTT.]]></title>
            <link>https://medium.com/@apshana/case-study-for-4-companies-on-big-data-uber-equifax-walmart-marriott-5c1e3eb6e6ed?source=rss-545dd27682cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/5c1e3eb6e6ed</guid>
            <dc:creator><![CDATA[Apshana S P]]></dc:creator>
            <pubDate>Sat, 15 Feb 2025 15:13:52 GMT</pubDate>
            <atom:updated>2025-02-15T15:13:52.141Z</atom:updated>
            <content:encoded><![CDATA[<h3>CASE STUDY FOR 4 COMPANIES ON BIG DATA : UBER , EQUIFAX , WALMART,MARRIOTT.</h3><h3>1. Uber — The Data Breach Cover-Up That Shook Trust</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/440/1*GB-qArNAxCb7PfzU3KMt3Q.png" /></figure><h3>About the Company</h3><p>Uber, the global ride-hailing giant, has transformed the transportation sector with its data-based business model. With presence in more than 70 countries, Uber depends heavily on big data to efficiently match riders with drivers.</p><h3>The Problem</h3><p>In 2016, Uber’s database was breached, revealing the personal information of 57 million drivers and users. Rather than disclosing the breach, Uber paid hackers $100,000 to remain silent, breaking legal and ethical requirements. The company was subsequently sued and ordered to pay a $148 million settlement.</p><h3>How Big Data Helped Solve It</h3><p>To restore trust and enhance security, Uber used big data analytics and AI-based cybersecurity methods, such as:</p><ul><li><strong>Real-time anomaly detection: </strong>Artificial intelligence now highlights unusual login efforts and unauthorized activity in Uber’s system.</li><li><strong>Predictive analytics to prevent fraud: </strong>Algorithms detect dubious patterns and forestall prospective cyber attacks.</li><li><strong>Encryption and tokenization:</strong> Personalized customer and driver information is encrypted now, diminishing exposure risk.</li></ul><h3>Lesson Learned</h3><p>Transparency is key. Uber’s case highlights the importance of <strong>ethical data governance</strong> and real-time security monitoring in big data operations.</p><h3>2. Equifax — The Largest Identity Theft Crisis in History</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IcnQ0D3Hn0mumT96ApOOfw.jpeg" /></figure><h3>About the Company</h3><p>Equifax is one of the big three U.S. credit reporting agencies that processes sensitive financial information for millions. Its infrastructure based on big data assists lenders and banks with creditworthiness assessment.</p><h3>The Problem</h3><p>Equifax experienced a humongous data breach for 147 million users in 2017 owing to an unpunched security bug. Hackers breached Social Security numbers, dates of birth, and financial information, leading to mass identity theft.</p><h3>How Big Data Helped Solve It</h3><p>Following the breach, Equifax applied sophisticated big data technologies to increase security and block future attacks:</p><ul><li><strong>Automated patching technology:</strong> Providing timely updates to software vulnerabilities.</li><li><strong>AI-driven risk scoring: </strong>Machine learning algorithms now evaluate vulnerabilities and notify security personnel in real time.</li><li><strong>Blockchain integration for safe identity verification:</strong> Rendering credit reports more tamper-proof and less susceptible to manipulation.</li></ul><h3>Lesson Learned</h3><p>Big data security must never be an afterthought. Businesses dealing with sensitive information have to focus on proactive cybersecurity protocols.</p><h3>3. Walmart — The Challenges of Managing Massive Supply Chain Data</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*lR1wDpc0EfqLgXtUD_Tsnw.jpeg" /></figure><h3>About the Company</h3><p>Walmart is the largest retailer globally, dealing with millions of customers every day in thousands of stores. Its supply chain processes depend greatly on big data for logistics optimization, inventory management, and customer demand prediction.</p><h3>The Problem</h3><p>With more than 2.5 petabytes of data created each hour, Walmart was having trouble effectively analyzing and leveraging this enormous amount of information. The company was plagued by:</p><ul><li><strong>Mismanaged inventory</strong> causing overstocking or stockouts.</li><li><strong>Supply chain</strong> inefficiencies resulting from poor demand forecasting</li><li><strong>Data silos</strong> leading to fragmented and delayed decision-making.</li></ul><h3>How Big Data Helped Solve It</h3><p>Walmart used big data analytics to transform its operations:<br>Real-time inventory tracking:</p><ul><li><strong>Real-time inventory tracking:</strong> AI-based demand forecasting: Forecasting customer demand based on buying habits and external influences such as weather and public holidays.</li><li><strong>AI-based demand forecasting:</strong> Forecasting customer demand based on buying habits and external influences such as weather and public holidays.</li><li><strong>Automated supply chain management: </strong>Using machine learning to optimize logistics and lower transportation expense and delivery time.</li></ul><h3>Lesson Learned</h3><p>Big data is a game-changer for large-scale operations. Centralized data integration and real-time analytics are essential to enhancing efficiency and decision-making.</p><h3>4. Marriott — The Hospitality Giant’s Cybersecurity Nightmare</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*x2y6Cs6rZner0Tjhr5ctzQ.png" /></figure><h3>About the Company</h3><p>Marriott is a global hotel giant with thousands of hotels spread across various brands. Big data assists it in tailoring customer experiences, optimizing prices, and streamlining operations.</p><h3>The Problem</h3><p>In <strong>2018, Marriott reported a hack on 500 million</strong> guests that compromised passport numbers, credit card information, and personal information. The hack lasted for four years, and there are questions raised about the security practices of the company.</p><h3>How Big Data Helped Solve It</h3><p>Marriott reacted by increasing its big data-based security protocols, including:</p><ul><li><strong>AI-driven intrusion detection:</strong> Live monitoring of guest information for suspicious activity.</li><li><strong>Behavioral analysis:</strong> Detection of suspicious transactions and unauthorized access.</li><li><strong>Secure cloud migration: </strong>Transferring information to secure cloud platforms with strong encryption.</li></ul><h3>Lesson Learned</h3><p>Cybersecurity has to be an <strong>ongoing process</strong>. Periodic audits and AI-based monitoring can avert breaches before they get out of hand.</p><h3>Conclusion — The Power and Pitfalls of Big Data</h3><p>Big data is a <strong>double sword.</strong> It facilitates potent insights and optimizations, but if data is not managed well, then it can cause devastating failures. The important takeaways from these case studies are:</p><ul><li><strong>Security must be proactive, not reactive</strong>. Companies must anticipate threats and act before breaches occur.</li><li><strong>AI and machine learning </strong>are essential for monitoring and risk detection. Predictive analytics can help prevent fraud, cyberattacks, and data mismanagement.</li><li><strong>Data quality matters as much as quantity.</strong> Biased or noisy data can lead to misleading conclusions, as seen in past cases.</li></ul><p>Big data is a game-changer, but it demands <strong>responsibility, transparency, and innovation</strong> to avoid becoming a liability. Companies that <strong>invest in ethical data practices and robust security systems</strong> will be the ones that thrive in the digital era.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5c1e3eb6e6ed" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ENTITY-RELATIONSHIP DIAGRAM]]></title>
            <link>https://medium.com/@apshana/entity-relationship-diagram-a8ab2158df8f?source=rss-545dd27682cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/a8ab2158df8f</guid>
            <category><![CDATA[entity-relationship-model]]></category>
            <category><![CDATA[real-time-application]]></category>
            <category><![CDATA[dbms]]></category>
            <dc:creator><![CDATA[Apshana S P]]></dc:creator>
            <pubDate>Sat, 10 Feb 2024 09:41:19 GMT</pubDate>
            <atom:updated>2024-02-10T09:43:23.658Z</atom:updated>
            <content:encoded><![CDATA[<h3>1) PHOTO-SHARING PLATFORM:</h3><p>Creating an Entity-Relationship (ER) diagram for a Photo Sharing Platform involves identifying the main entities, their attributes, and the relationships between them. Here’s a basic ER diagram for a photo sharing platform:</p><p>Entities:<br>1. User:<br> — UserID (Primary Key)<br> — Username<br> — Email<br> — Password<br> — Profile Picture<br> — Bio<br> — Registration Date</p><p>2. Photo:<br> — PhotoID (Primary Key)<br> — UserID (Foreign Key)<br> — Image URL<br> — Description<br> — Upload Date<br> — Likes<br> — Comments</p><p>3. Album:<br> — AlbumID (Primary Key)<br> — UserID (Foreign Key)<br> — Title<br> — Description<br> — Creation Date</p><p>Relationships:<br>1. User-Photo:<br> — One user can upload many photos.<br> — One photo is uploaded by one user.<br> — Relationship Type: One-to-Many<br> — Foreign Key: UserID in Photo table referencing UserID in User table.</p><p>2. User-Album:<br> — One user can create many albums.<br> — One album is created by one user.<br> — Relationship Type: One-to-Many<br> — Foreign Key: UserID in Album table referencing UserID in User table.</p><p>3. Photo-Album (Many-to-Many):<br> — A photo can belong to many albums.<br> — An album can contain many photos.<br> — Relationship Type: Many-to-Many<br> — Association Table: PhotoAlbum<br> — Foreign Keys:<br> — PhotoID in PhotoAlbum table referencing PhotoID in Photo table.<br> — AlbumID in PhotoAlbum table referencing AlbumID in Album table.</p><p>This ER diagram captures the basic structure of a photo sharing platform, allowing users to upload photos, create albums, like photos, and comment on them. It also supports the relationship between users and their content, as well as the association between photos and albums.</p><h3>E-R DIAGRAM:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3YcH1VZ3xL9ddgcEYM-_jw.png" /><figcaption>PHOTO-SHARING PLATFORM</figcaption></figure><h3>2) PUBLIC HEALTH INFORMATION SYSTEM:</h3><p>Creating an ER diagram for a Public Health Information System involves identifying the main entities related to public health data management.</p><p>Entities:</p><p>1. Patient:</p><p>- PatientID (Primary Key),Name,Date of Birth,Gender,Address,Phone Number,Email</p><p>2. Medical Record:</p><p>- RecordID (Primary Key),PatientID (Foreign Key),Admission Date,Discharge Date,Diagnosis,Treatment,Medications,Lab Results,Imaging Results,PhysicianID</p><p>3. Physician:</p><p>- PhysicianID (Primary Key), Name, Specialty,Contact Information</p><p>4. Public Health Data:</p><p>- ReportID (Primary Key)</p><p>- Date,Disease/Condition,Number of Cases,Location, Source</p><p>Relationships:</p><p>1. Patient-Medical Record:</p><p>- One patient can have many medical records.</p><p>- Each medical record belongs to one patient.</p><p>- Relationship Type: One-to-Many</p><p>- Foreign Key: PatientID in Medical Record table referencing PatientID in Patient table.</p><p>2. Physician-Medical Record:</p><p>One physician can be associated with many medical records.</p><p>- Each medical record is associated with one physician.</p><p>- Relationship Type: One-to-Many</p><p>- Foreign Key: PhysicianID in Medical Record table referencing PhysicianID in Physician table.</p><p>3. Public Health Data-Medical Record (Many-to-Many):</p><p>- A medical record may contribute to public health data.</p><p>- Public health data may involve multiple medical records.</p><p>- Relationship Type: Many-to-Many</p><p>- Association Table: RecordData</p><p>- Foreign Keys:</p><p>- RecordID in RecordData table referencing RecordID in Medical Record table.</p><p>- ReportID in RecordData table referencing ReportID in Public Health Data table.</p><p>This ER diagram provides a basic structure for a Public Health Information System, facilitating the management of patient records, physician information, and public health data.</p><h3>E-R DIAGRAM:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Bug_gR9X3-mppllavZpBpA.png" /><figcaption>PUBLIC HEALTH INFORMATION SYSTEM</figcaption></figure><h3>3) HOMELESS SHELTHER MANAGEMENT SYSTEM:</h3><p>Designing an ER (Entity-Relationship) diagram for a Homeless Shelter Management System involves identifying the main entities, their attributes, and the relationships between them. Here’s a simplified example:</p><p>Entities:<br>1. Shelter<br>2. Resident<br>3. Staff<br>4. Donor<br>5. Donation<br>6. Service<br>7. Bed</p><p>Attributes:<br>1. Shelter:<br> — Shelter_ID (Primary Key)<br> — Name<br> — Location<br> — Capacity<br> — Phone_Number</p><p>2. Resident:<br> — Resident_ID (Primary Key)<br> — Name<br> — Age<br> — Gender<br> — Entry_Date<br> — Exit_Date<br> — Reason_For_Homelessness<br> — Shelter_ID (Foreign Key)</p><p>3. Staff:<br> — Staff_ID (Primary Key)<br> — Name<br> — Position<br> — Contact_Info<br> — Shelter_ID (Foreign Key)</p><p>4. Donor:<br> — Donor_ID (Primary Key)<br> — Name<br> — Contact_Info</p><p>5. Donation:<br> — Donation_ID (Primary Key)<br> — Amount<br> — Donation_Date<br> — Donor_ID (Foreign Key)</p><p>6. Service:<br> — Service_ID (Primary Key)<br> — Type<br> — Description<br> — Date<br> — Shelter_ID (Foreign Key)</p><p>7. Bed:<br> — Bed_ID (Primary Key)<br> — Status (e.g., occupied, available)<br> — Shelter_ID (Foreign Key)</p><p>Relationships:<br>1. Shelter-Resident: One shelter can have many residents. One resident belongs to one shelter.<br>2. Shelter-Staff: One shelter can have many staff members. One staff member works at one shelter.<br>3. Donor-Donation: One donor can make many donations. One donation is made by one donor.<br>4. Shelter-Donation: One shelter can receive many donations. One donation is received by one shelter.<br>5. Shelter-Service: One shelter can provide many services. One service is provided by one shelter.<br>6. Shelter-Bed: One shelter can have many beds. One bed belongs to one shelter.</p><p>This ER diagram outlines the basic structure and relationships of a Homeless Shelter Management System.</p><h3>E-R DIAGRAM:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2KPAXpa9ztifSqutRFVWzw.png" /><figcaption>HOMELESS SHELTHER MANAGEMENT SYSTEM</figcaption></figure><h3>4) TRAFFIC MANAGEMENT SYSTEM:</h3><p>A Traffic Management System is a critical component of urban infrastructure, aiming to optimize traffic flow, enhance safety, and streamline enforcement efforts. Designing an effective system requires careful consideration of various entities, their attributes, and the relationships between them. This writeup discusses the process of designing an Entity-Relationship (ER) diagram for a Traffic Management System.</p><p>Entities and Attributes:</p><p>1. Vehicle:</p><p>- Attributes: vehicle_id(PK), license_plate, make, model, year, owner_name</p><p>2. Traffic Incident:</p><p>- Attributes: incident_id(PK), date_time, vehicle_id(FK), officer_id, severity, description</p><p>3. Intersection:</p><p>- Attributes: intersection_id(PK), location, traffic_light_id(FK)</p><p>4. Traffic Light:</p><p>- Attributes: traffic light_id(PK), intersection_id(FK), status</p><p>5. Ticket:</p><p>- Attributes: ticket_id(PK), violation_date, violation_type, fine_amount, vehicle_id(FK), officer_id(FK), incident_id(FK)</p><p>Relationships:</p><p>1. Vehicle — Traffic Incident:</p><p>- Relationship: A Vehicle can be involved in multiple Traffic Incidents.</p><p>- Multiplicity: One-to-Many (1:N)</p><p>2. Traffic Incident — Intersection:</p><p>- Relationship: Traffic Incidents occur at specific Intersections.</p><p>- Multiplicity: One-to-One (1:1)</p><p>3. Traffic Light — Intersection:</p><p>- Relationship: Traffic Lights are installed at Intersections.</p><p>- Multiplicity: One-to-One (1:1)</p><p>4. Ticket — Vehicle:</p><p>- Relationship: Tickets are issued to specific Vehicles.</p><p>- Multiplicity: One-to-Many (1:N)</p><p>5. Ticket — Traffic Incident:</p><p>- Relationship: Tickets are associated with Traffic Incidents.</p><p>- Multiplicity: One-to-One (1:1)</p><p>The ER diagram provides a structured representation of entities, attributes, and relationships within a Traffic Management System. It serves as a blueprint for database design and system development, facilitating efficient traffic monitoring, incident management, and enforcement activities.</p><h3>E-R DIAGRAM:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ebpy4evZYMzOZK412VOPoA.png" /><figcaption>HOMELESS SHELTHER MANAGEMENT SYSTEM</figcaption></figure><h3>5)MUSEUM EXHIBITION MANAGEMENT SYSTEM:</h3><p>A Museum Exhibition Management System is essential for organizing, cataloging, and presenting exhibitions within a museum. This system facilitates the efficient management of artworks, artists, curators, and exhibitions. Designing an effective system requires identifying the key entities, their attributes, and the relationships between them.</p><p>Entities:</p><p>Exhibition:</p><p>exhibition_id (Primary Key), title,start_date,end_date,location,curator_id (Foreign Key referencing Curator)</p><p>Artwork:</p><p>artwork_id (Primary Key),title,artist_id (Foreign Key referencing Artist),medium,year_created</p><p>Artist:</p><p>artist_id (Primary Key),name,birth_date,nationality,style</p><p>Curator:</p><p>curator_id (Primary Key), name, specialty</p><p>Relationships:</p><p>Exhibition — Artwork:</p><p>- One Exhibition can feature many Artworks. (One-to-Many)</p><p>- Foreign key: exhibition_id in Artwork referencing Exhibition.</p><p>Artwork — Artist:</p><p>- Each Artwork is created by one Artist. (One-to-One)</p><p>Foreign key: artist_id in Artwork referencing Artist.</p><p>Exhibition — Curator:</p><p>- Each Exhibition is curated by one Curator. (One-to-One)</p><p>Foreign key: curator_id in Exhibition referencing Curator.</p><p>The ER diagram provides a comprehensive view of the entities, attributes, and relationships within a Museum Exhibition Management System. It serves as a blueprint for database design and system development, facilitating efficient organization, cataloging, and presentation of museum exhibitions.</p><h3>E-R DIAGRAM:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*86bvCwg7-yjlMn4WT644wQ.png" /><figcaption>MUSEUM EXHIBITION MANAGEMENT SYSTEM</figcaption></figure><p>— — — — — — — — — — — — — — END OF THE PAGE— — — — — — — — — — — — — —</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a8ab2158df8f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[PREDICTING USERS EMOTIONAL INTELLIGENCE BASED ON SOCIAL NETWORKING DATA]]></title>
            <link>https://medium.com/@apshana/predicting-users-emotional-intelligence-based-on-social-networking-data-9a988ea81e15?source=rss-545dd27682cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/9a988ea81e15</guid>
            <category><![CDATA[case-study]]></category>
            <category><![CDATA[predictive-analytics]]></category>
            <category><![CDATA[social-networking]]></category>
            <dc:creator><![CDATA[Apshana S P]]></dc:creator>
            <pubDate>Fri, 26 Jan 2024 11:29:04 GMT</pubDate>
            <atom:updated>2024-01-26T12:16:06.012Z</atom:updated>
            <content:encoded><![CDATA[<p><em>PREDICTING USERS EMOTIONAL INTELLIGENCE BASED ON SOCIAL NETWORKING DATA | by Apshana S P</em></p><h3>INTRODUCTION:</h3><p>Social networks have integrated into daily lives of most people in the way of interactions and of lifestyles. The users’ identity, relationships, or other characteristics can be explored from the social networking data, in order to provide more personalized services to the users. In this work, we focus on predicting the user’s emotional intelligence (EI) based on the social networking data. As an essential facet of users’ psychological characteristics, EI plays an important role on well-being, interpersonal relationships, and overall success in people’s life. Most existing work on predicting users’ emotional intelligence is based on questionnaires that may collect dishonest answers or unconscientious responses, thus leading in potentially inaccurate prediction results. In this work, we are motivated to propose an emotional intelligence prediction model based on the sentiment analysis of social networking data. The model is represented by four dimensions including self-awareness, self-regulation, self-motivation and social relationships. The EI of a user is then measured by the four numerical values or the sum of them. In the experiments, we predict the EIs of over a hundred thousand users based on one of the largest social networks of China, Weibo.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kjm8926FfhTA49m-bl43ug.jpeg" /></figure><h3>DATA DESCRIPTION AND PREPROCESSING:</h3><p>Creating a dataset for predicting emotional intelligence (EI) based on social networking data involves incorporating various features that may influence or reflect emotional intelligence. Here’s a simplified and hypothetical dataset for illustrative purposes:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/665/1*LrZ-4wxQwM8p4WqQnivuaA.png" /></figure><p>These numerical values could be derived from sentiment analysis of the user’s posts, interactions, comments, and other activities on a social network platform. Natural Language Processing (NLP) techniques can be employed to analyze and quantify the sentiment expressed in the textual data. Each dimension represents a specific aspect of emotional intelligence, and the sum provides an aggregate score.</p><h3>ALGORITHMS USED AND REASONING:</h3><p>To predict emotional intelligence (EI) scores based on the provided dataset using Natural Language Processing (NLP), we can create a simple model that analyzes textual data associated with each user. Here’s a basic example using Python with the scikit-learn library for simplicity.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/533/1*qqlJcNmAoHbBitrIod5d2Q.png" /></figure><p>Mean Squared Error: 0.08</p><p>R-squared (R2) Score: 0.92</p><h3>RESULT:</h3><ul><li>actual values. A lower MSE indicates better model performance. In this case, a MSE of 0.06 suggests relatively accurate predictions.</li><li>R-squared (R2) Score: The proportion of the variance in the dependent variable (EI_Score) that is predictable from the independent variables. An R2 score of 0.92 indicates that 92% of the variability in EI scores is captured by the model, suggesting a good fit.</li></ul><h3>CONCLUSION:</h3><p>The linear regression model predicts Emotional Intelligence scores based on self-awareness, self-regulation, self-motivation, and social relationships. It has good predictive performance, with a low MSE and a strong correlation between selected features and users’ Emotional Intelligence. The model can be used to develop interventions to enhance emotional well-being and interpersonal relationships. However, improvements include using actual textual data, advanced Natural Language Processing techniques, and additional features. Ethical considerations and transparency are also crucial for responsible use of social networking data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*D6vBbUpp1Y1vv-1OctBUqA.jpeg" /></figure><h3>REFERENCES:</h3><p><a href="https://link.springer.com/">https://link.springer.com</a></p><p><a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1636947/">https://www.ncbi.nlm.nih.gov</a></p><p><a href="https://scienceforwork.com/en/blog/emotional-intelligence-performance/">https://scienceforwork.com</a></p><p>Quillbot and chatgpt for context reference</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9a988ea81e15" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[PREDICTIVE ANALYSIS IN E-TOURISM(predictive model to forecast whether a user will convert their…]]></title>
            <link>https://medium.com/@apshana/predictive-analysis-in-e-tourism-predictive-model-to-forecast-whether-a-user-will-convert-their-247757eb164f?source=rss-545dd27682cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/247757eb164f</guid>
            <category><![CDATA[predictive-analyses]]></category>
            <category><![CDATA[case-study]]></category>
            <category><![CDATA[tourisme]]></category>
            <dc:creator><![CDATA[Apshana S P]]></dc:creator>
            <pubDate>Fri, 26 Jan 2024 08:10:33 GMT</pubDate>
            <atom:updated>2024-01-26T11:03:58.553Z</atom:updated>
            <content:encoded><![CDATA[<h3>PREDICTIVE ANALYSIS IN E-TOURISM(predictive model to forecast whether a user will convert their interaction into a booking)</h3><p><em>PREDICTIVE ANALYSIS IN E-TOURISM(predictive model to forecast whether a user will convert their… | by Apshana S P</em></p><h3>INTRODUCTION:</h3><p>In 2018, travel and tourism was the second-fastest-growing sector, adding $8.8 trillion to the world’s GDP and 319 million mint-new jobs. People are travelling a lot more, and factors like social media and extended media coverage are increasing the reach to even the farthest corners of the world. As a result of this heightened awareness, there has been a veritable boom in various facets of the travel industry.</p><p>Technology and<a href="https://www.mygreatlearning.com/blog/use-of-big-data-analytics-in-everyday-life/"> big data</a> have been crucial players in this boom. Travel technology has seen its best decade so far, as have traditional players. Digital innovation and emerging technologies have helped entirely new segments to enter and revolutionise the travel market — from accommodation brands to solo travelling agencies and everything in between.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/980/1*ByJCoTp09EvwZ_gpO-x-XA.jpeg" /></figure><h3>DATA DESCRIPTION AND PRE PROCESSING:</h3><p>A detailed dataset for a predictive analytics case study in e-tourism involves various parameters. Below is a simplified and hypothetical dataset with some key variables that could be considered for analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/683/1*ICB6YScpi6yKEhP_23DVtw.png" /></figure><p>In a predictive analytics scenario for the e-tourism dataset provided, let’s consider a simplified predictive model to forecast whether a user will convert their interaction into a booking (Converted_to_Booking). We&#39;ll use a hypothetical binary classification model for this example, and logistic regression can be a suitable algorithm for this type of prediction.</p><h3>ALGORITHMS USED AND REASONING:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/537/1*r0lLXTzOELSqEkuWAnBD0Q.png" /></figure><p>It splits the data into training and testing sets, creates a logistic regression model, trains the model, makes predictions on the test set, and evaluates the model’s performance.</p><p>Accuracy: 0.85</p><p>Confusion Matrix: [[120 10] [ 25 145]]</p><ul><li>Accuracy: 85% indicates the proportion of correctly predicted outcomes.</li><li>Confusion Matrix: Shows the number of true positive, true negative, false positive, and false negative predictions.</li><li>Classification Report: Provides precision, recall, and F1-score for each class (0: Not Converted, 1: Converted), along with support (number of instances).</li></ul><p>In this hypothetical output:</p><ul><li>Precision for class 0: 83% (TP / (TP + FP))</li><li>Recall for class 0: 92% (TP / (TP + FN))</li><li>F1-score for class 0: 87%</li></ul><p>These metrics give insights into how well the predictive model is performing in distinguishing between users who convert interactions into bookings and those who do not.</p><h3>CONCLUSION:</h3><p>The application of predictive analytics in e-tourism has shown promise, with a model achieving 85% accuracy in predicting user conversion. The model’s accuracy was attributed to features like Stay Duration, Hotel Rating, Price per Night, and Website Interactions. The model also performed well in predicting both positive and negative instances, with a precision of 83% and recall of 92%. The F1-score of 87% reflects a good balance between precision and recall. The model’s success has practical implications for e-tourism platforms, allowing businesses to identify potential customers, tailor marketing strategies, personalize user experiences, and optimize pricing.</p><h3>REFERENCES:</h3><p><a href="https://www.springbord.com/">https://www.springbord.com</a></p><p><a href="https://link.springer.com/10.1007/978-3-030-05324-6_29-1">https://link.springer.com</a></p><p>Quillbot,chatgpt for context reference</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=247757eb164f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[PREDICTIVE ANALYSIS FOR CLASSIFICATION OF DRY BEANS BY ITS FEAUTURES]]></title>
            <link>https://medium.com/@apshana/predictie-analysis-for-agricultural-data-7481346f62c3?source=rss-545dd27682cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/7481346f62c3</guid>
            <category><![CDATA[agriculture]]></category>
            <category><![CDATA[case-study-assignment]]></category>
            <category><![CDATA[image-processing]]></category>
            <dc:creator><![CDATA[Apshana S P]]></dc:creator>
            <pubDate>Thu, 25 Jan 2024 20:53:34 GMT</pubDate>
            <atom:updated>2024-01-26T12:27:54.693Z</atom:updated>
            <content:encoded><![CDATA[<p><em>PREDICTIVE ANALYSIS FOR AGRICULTURAL DATA | by Apshana S P | Medium</em></p><h3>INTRODUCTION:</h3><p>Predictive analytics has revolutionized agriculture by providing farmers with insights and decision-making capabilities. This shift towards data-driven agriculture optimizes resource allocation, improves crop yields, and enhances sustainability. A forward-thinking farming cooperative, used predictive analytics to predict and optimize key parameters, transforming their operations into more precise, efficient, and economically viable endeavors. This case study highlights how predictive analytics can empower farmers to make informed decisions, adapt to dynamic conditions, and contribute to a sustainable and profitable future for the agriculture industry.</p><h3>DATA DESCRIPTION AND PRE PROCESSING:</h3><p>Seven different types of dry beans were used in this research, taking into account the features such as form, shape, type, and structure by the market situation. A computer vision system was developed to distinguish seven different registered varieties of dry beans with similar features in order to obtain uniform seed classification. For the classification model, images of 13,611 grains of 7 different registered dry beans were taken with a high-resolution camera. Bean images obtained by computer vision system were subjected to segmentation and feature extraction stages, and a total of 16 features; 12 dimensions and 4 shape forms, were obtained from the grains</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/953/1*i7s8poM-vhuF3HAp8a_BFA.png" /></figure><p>The aim of this study was to develop an artificial intelligence-based CVS for the classification of basic types of dry beans determined by the Turkish Standards Institute, which contains morphologically similar features and has no distinctive color features.</p><h3>ALGORITHMS USED AND REASONING:</h3><p>In this study, firstly, the images of dry beans were obtained, and unwanted noise on the image was cleared by image processing methods. Histogram information of gray image information was obtained, and histogram equalization was made. In the image where beans became clear, dry beans were thresholded by Otsu’s method (Otsu, 1979). The Otsu’s algorithm is based on calculating the optimal threshold value by processing the histogram information of the image.</p><p>Otsu’s method is commonly used for automatic image thresholding.Here’s a general process using Python and scikit-image to apply the Otsu algorithm and perform image segmentation Of the dry bean :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/863/1*r0MyCqjXVPTRH8SQrjjXgQ.png" /></figure><h3><strong>OUTPUT AND RESULT:</strong></h3><p>The dataset demonstrating a comprehensive characterization of bean seeds. The Otsu algorithm is used for segmentation or thresholding, enhancing efficiency and accuracy. This approach aligns with the trend of leveraging computational methods for agricultural applications. The dataset suggests a multidisciplinary approach to bean seed analysis, with machine learning potentially enhancing crop monitoring and quality assessment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/253/1*eQIHB76RWHAcuhCjg6l24A.png" /><figcaption>OUTPUT</figcaption></figure><h3>REFERENCES:</h3><ol><li>insights from the article by KOKLU, M. and OZKAN, I.A., (2020), “Multiclass Classification of Dry Beans Using Computer Vision and Machine Learning Techniques.” <em>Computers and Electronics in Agriculture</em>, 174, 105507.</li><li>chatgpt to rephrase some of the context.</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7481346f62c3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[PREDICTIE ANALYSIS FOR SMART TRAFFIC MANAGEMENT]]></title>
            <link>https://medium.com/@apshana/predictie-analysis-for-smart-traffic-management-b2ddc1020959?source=rss-545dd27682cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/b2ddc1020959</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[predictive-analyses]]></category>
            <category><![CDATA[paper]]></category>
            <dc:creator><![CDATA[Apshana S P]]></dc:creator>
            <pubDate>Thu, 25 Jan 2024 19:10:46 GMT</pubDate>
            <atom:updated>2024-01-26T12:28:39.204Z</atom:updated>
            <content:encoded><![CDATA[<p><em>PREDICTIE ANALYSIS FOR SMART TRAFFIC MANAGEMENT | by Apshana S P | Medium</em></p><h3>INTRODUCTION:</h3><p>Smart Traffic management system (STMS) is a one of the important feature for smart city. Currently traffic management and alert systems are not fulling needs of STMS. It is more expensive and highly configurable to provide better service for traffic management. Predictive analytics in smart traffic management involves forecasting future traffic conditions based on historical data, real-time inputs, and various influencing factors.By using predictive analysis for smart traffic management we can</p><p>. <strong>Reduce traffic congestion and commute times.</strong></p><p><strong>.Enhance overall traffic flow and minimize bottlenecks.</strong></p><p>. <strong>Improve air quality and reduce environmental impact.</strong></p><p>. <strong>Optimize resource allocation and city planning based on data insights.</strong></p><p>There are several analytical scriptures to analyze the traffic density and provide solution through predictive analytics.let us see some the the predictions used for STMS in this case study and gather and understand insights of it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*l8XG9jpcg-E7tht36n6BZw.jpeg" /></figure><h3>ALGORITHMS USED FOR PREDICTION:</h3><p>· Predictive analysis is used by traffic planners to forecast future conditions so that personnel are better able to deal with it effectively in terms of resource allocation, route optimization to minimize traffic congestion, and the adjustment of traffic signal times. Traffic prediction models are models that forecast traffic conditions using techniques trained to extract relevant information from historical or real-time data about traffic flow, congestion levels, commute times, etc. and use it to generate accurate and reliable predictions.</p><p>· We can manage traffic also by analyzing the previous traffic data of a particular route and predict traffic at that given time .</p><p>· We may also take a view at the speed limit of the route and the total vechicles travelling in that route to predict traffic and imform the traveler earlierto take some other route to prevent traffic.</p><p>. Traffic can also be prevented by notifying if any accidents took place during the time of day or any accident prone zones and predict the traffic around the accident area and manage it.</p><h3>DATA REQUIRED FOR PREDICTION:</h3><ul><li>Current Weather Conditions: Real-time information about weather conditions, including rain, snow, storms, and temperature.</li><li>Weather Forecasts: Predictions for upcoming weather conditions that may impact traffic.</li><li>Event Schedules: Information about scheduled events such as sports games, concerts, or parades.</li><li>Expected Attendance: Estimates of the number of people attending special events.</li><li>Parking Space Availability: Real-time information on available parking spaces in different areas.</li><li>Parking Regulations: Rules and regulations related to parking.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/685/1*rp2HX0wRMkFuyvbZbGRZyw.png" /></figure><h3>CONCLUSION:</h3><p>Predictive analytics in smart traffic management systems offers numerous benefits, including proactive traffic management, optimized resource allocation, enhanced public safety, efficient route guidance, environmental impact reduction, data-driven decision-making, and improved quality of life. By forecasting future traffic conditions, cities can mitigate congestion, reduce commute times, and enhance traffic flow. The integration of predictive models and technology ensures cities can adapt to urban mobility dynamics.</p><h3>REFERENCES:</h3><ol><li><a href="http://www.isarsoft.com/">www.isarsoft.com</a></li><li><a href="http://www.sciencedirect.com/">www.sciencedirect.com</a></li><li><a href="http://www.researchgate.net/">www.researchgate.net</a></li><li>Quillbot,chatgpt for abstraction</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b2ddc1020959" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>