https://iftekhairul.github.io/Muhammad Ifte Khairul IslamWowchemy (https://wowchemy.com)en-usSun, 15 Dec 2024 02:33:00 +0000https://iftekhairul.github.io/media/icon_hu0b7a4cb9992c9ac0e91bd28ffd38dd00_9727_512x512_fill_lanczos_center_3.pnghttps://iftekhairul.github.io/https://iftekhairul.github.io/publication/dygcl/Sun, 15 Dec 2024 02:33:00 +0000https://iftekhairul.github.io/publication/dygcl/<p>Predicting events, ranging from political unrest to disease outbreaks and criminal activities, stands as a pivotal task in proactively addressing emerging challenges. Despite the richness of textual data as a source for event detection, it is challenging to extract contextual information from documents due to their complex structure and the dynamic evolution of events. In response to this challenge, dynamic Graph Neural Networks (GNNs) have emerged as a promising tool for capturing the intricate patterns embedded within textual data graphs. Nevertheless, many models in this domain primarily rely on local node-level representations, overlooking the essential global graph-level context. However, both node-level and graph-level representations are critical for effective event prediction. Node-level representations provide insight into the local structure, while graph-level representations offer an understanding of the global structure and the evaluation of temporal graphs. To address these challenges, in this paper, we propose a Dynamic Graph Contrastive Learning (DyGCL) method for event prediction. Our model DyGCL first employs a local view encoder to effectively capture the local dynamic structure of input graphs as the evolving node representations. Then, it performs a global view encoder to perceive the hierarchical dynamic graph representation of the input graphs. Finally, the graph representations from both encoders, optimized via contrastive learning, are combined with an attention mechanism and utilized to predict future events. Our extensive experiments demonstrate that our proposed method outperforms the state-of-the-art methods for event prediction on six real-world datasets.</p>https://iftekhairul.github.io/publication/mpool/Thu, 25 May 2023 02:33:00 +0000https://iftekhairul.github.io/publication/mpool/<p>Recently, Graph Neural Networks (GNNs) have emerged as a powerful technique for various graph-related tasks. Current GNN models apply different graph pooling methods that reduce the number of nodes and edges to learn the higher-order structure of the graph in a hierarchical way. However, these methods primarily rely on the one-hop neighborhood and do not consider the higher-order structure of the graph. To address this issue, in this work, we propose a multi-channel Motif-based Graph Pooling method named (MPool) that captures the higher-order graph structure with motif and also considers the local and global graph structure through a combination of selection and clustering-based pooling operations. In the first channel, we develop node selection-based graph pooling by designing a node ranking model considering the motif adjacency of nodes. In the second channel, we develop cluster-based graph pooling by designing a spectral clustering model using motif adjacency. Finally, the result of each channel is aggregated into the final graph representation. We perform extensive experiments and demonstrate that our proposed method outperforms the baseline methods for graph classification tasks on eight benchmark datasets.</p>https://iftekhairul.github.io/publication/seqhygan/Tue, 09 May 2023 02:33:00 +0000https://iftekhairul.github.io/publication/seqhygan/<p>Extracting meaningful features from sequences and devising effective similarity measures are vital for sequence data mining tasks, particularly sequence classification. While neural network models are commonly used to automatically learn sequence features, they are limited to capturing adjacent structural connection information and ignoring global, higher-order information between the sequences. To address these challenges, we propose a novel Hypergraph Attention Network model, namely Seq-HyGAN for sequence classification problems. To capture the complex structural similarity between sequence data, we create a novel hypergraph model by defining higher-order relations between subsequences extracted from sequences. Subsequently, we introduce a Sequence Hypergraph Attention Network that learns sequence features by considering the significance of subsequences and sequences to one another. Through extensive experiments, we demonstrate the effectiveness of our proposed Seq-HyGAN model in accurately classifying sequence data, outperforming several state-of-the-art methods by a significant margin.</p>https://iftekhairul.github.io/post/working-as-sub-reviewer-for-complex-networks-2021/Thu, 16 Sep 2021 19:55:38 +0000https://iftekhairul.github.io/post/working-as-sub-reviewer-for-complex-networks-2021/<p>I have the opportunity to work as sub-reviewer for Complex Networks 2021. I will earnestly try to review paper with my knowledge and skillset.</p>https://iftekhairul.github.io/publication/example/Sun, 12 Sep 2021 02:33:00 +0000https://iftekhairul.github.io/publication/example/<p>Drug-drug interaction (DDI) indicates the event where a particular drug’s desired course of action is modified when taken together with other drugs (s). DDIs may hamper, enhance, or reduce the expected effect of either drug or, at the worst possible scenario, cause an adverse side effect. While it is crucial to identify drug-drug interactions, it is quite impossible to detect all possible DDIs for a new drug during the clinical trial. Therefore, many computational methods are proposed for this task. In this paper, we propose a novel method, HIN-DDI for discovering DDIs. This method considers drugs and other biomedical entities like proteins, pathways, and side effects, for DDI prediction. We design a heterogeneous information network (HIN) to model relations between these entities. Afterward, we extract the rich semantic relationships among these entities using different meta-path-based topological features. An extensive set of features are fed to different classifiers for DDI prediction. Moreover, we run extensive experiments to compare and evaluate the effectiveness of HIN-DDI with other methods. Results exhibit that HIN-DDI is quite effective in predicting new drugs as well as existing drugs. Unlike existing works, HIN-DDI can predict new drugs, and more importantly, it can impressively outmatch baseline methods by up to 63%.</p>https://iftekhairul.github.io/publication/proximity-based-compression-for-network-embedding-3-2020-n-pag/Sun, 12 Sep 2021 02:20:48 +0000https://iftekhairul.github.io/publication/proximity-based-compression-for-network-embedding-3-2020-n-pag/<p>Network embedding that encodes structural information of graphs into a low-dimensional vector space has been proven to be essential for network analysis applications, including node classification and community detection. Although recent methods show promising performance for various applications, graph embedding still has some challenges; either the huge size of graphs may hinder a direct application of the existing network embedding method to them, or they suffer compromises in accuracy from locality and noise. In this paper, we propose a novel Network Embedding method, NECL, to generate embedding more efficiently or effectively. Our goal is to answer the following two questions: 1) Does the network Compression significantly boost Learning? 2) Does network compression improve the quality of the representation? For these goals, first, we propose a novel graph compression method based on the neighborhood similarity that compresses the input graph to a smaller graph with incorporating local proximity of its vertices into super- nodes; second, we employ the compressed graph for network embedding instead of the original large graph to bring down the embedding cost and also to capture the global structure of the original graph; third, we refine the embeddings from the compressed graph to the original graph. NECL is a general meta-strategy that improves the efficiency and effectiveness of many state-of-the-art graph embedding algorithms based on node proximity, including DeepWalk, Node2vec, and LINE. Extensive experiments validate the efficiency and effectiveness of our method, which decreases embedding time and improves classification accuracy as evaluated on single and multi-label classification tasks with large real-world graphs.</p>https://iftekhairul.github.io/post/paper-accepted-in-cibcb-2021/Sun, 12 Sep 2021 02:03:52 +0000https://iftekhairul.github.io/post/paper-accepted-in-cibcb-2021/<p>I am glad to inform that our research paper titled &ldquo;Predicting Drug-Drug Interactions Using Meta-path Based Similarities&rdquo; has been accepted at IEEE International Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB) 2021.</p>https://iftekhairul.github.io/post/working-as-co-ordinator-for-reu-site-big-data-analytics-at-oklahoma-state-university/Tue, 01 Jun 2021 19:51:39 +0000https://iftekhairul.github.io/post/working-as-co-ordinator-for-reu-site-big-data-analytics-at-oklahoma-state-university/<p>I have the opportunity to co-ordinate the program for <strong>REU Site: Big Data Analytics at Oklahoma State University.</strong> I hold daily meetings with 10 undergraduate students from 5 different universities, keeping track of their progress and helping them out in their projects. Moreover, I am actively involved in undergraduate students' projects.</p>https://iftekhairul.github.io/post/working-as-a-judge-oklahoma-state-science-and-engineering-fair-2021/Thu, 01 Apr 2021 19:59:25 +0000https://iftekhairul.github.io/post/working-as-a-judge-oklahoma-state-science-and-engineering-fair-2021/<p>I have the pleasure to work as a part of judging judge in Oklahoma State Science and Engineering Fair (OSSEF) 2021 under &ldquo;<strong>Science and Engineering</strong>&rdquo; Category. I worked as a judge for OSSEF 2020 as well.</p>https://iftekhairul.github.io/slides/example/Tue, 05 Feb 2019 00:00:00 +0000https://iftekhairul.github.io/slides/example/<h1 id="create-slides-in-markdown-with-wowchemy">Create slides in Markdown with Wowchemy</h1> <p><a href="https://wowchemy.com/" target="_blank" rel="noopener">Wowchemy</a> | <a href="https://owchemy.com/docs/managing-content/#create-slides" target="_blank" rel="noopener">Documentation</a></p> <hr> <h2 id="features">Features</h2> <ul> <li>Efficiently write slides in Markdown</li> <li>3-in-1: Create, Present, and Publish your slides</li> <li>Supports speaker notes</li> <li>Mobile friendly slides</li> </ul> <hr> <h2 id="controls">Controls</h2> <ul> <li>Next: <code>Right Arrow</code> or <code>Space</code></li> <li>Previous: <code>Left Arrow</code></li> <li>Start: <code>Home</code></li> <li>Finish: <code>End</code></li> <li>Overview: <code>Esc</code></li> <li>Speaker notes: <code>S</code></li> <li>Fullscreen: <code>F</code></li> <li>Zoom: <code>Alt + Click</code></li> <li><a href="https://github.com/hakimel/reveal.js#pdf-export" target="_blank" rel="noopener">PDF Export</a>: <code>E</code></li> </ul> <hr> <h2 id="code-highlighting">Code Highlighting</h2> <p>Inline code: <code>variable</code></p> <p>Code block:</p> <pre><code class="language-python">porridge = &quot;blueberry&quot; if porridge == &quot;blueberry&quot;: print(&quot;Eating...&quot;) </code></pre> <hr> <h2 id="math">Math</h2> <p>In-line math: $x + y = z$</p> <p>Block math:</p> <p>$$ f\left( x \right) = ;\frac{{2\left( {x + 4} \right)\left( {x - 4} \right)}}{{\left( {x + 4} \right)\left( {x + 1} \right)}} $$</p> <hr> <h2 id="fragments">Fragments</h2> <p>Make content appear incrementally</p> <pre><code>{{% fragment %}} One {{% /fragment %}} {{% fragment %}} **Two** {{% /fragment %}} {{% fragment %}} Three {{% /fragment %}} </code></pre> <p>Press <code>Space</code> to play!</p> <span class="fragment " > One </span> <span class="fragment " > **Two** </span> <span class="fragment " > Three </span> <hr> <p>A fragment can accept two optional parameters:</p> <ul> <li><code>class</code>: use a custom style (requires definition in custom CSS)</li> <li><code>weight</code>: sets the order in which a fragment appears</li> </ul> <hr> <h2 id="speaker-notes">Speaker Notes</h2> <p>Add speaker notes to your presentation</p> <pre><code class="language-markdown">{{% speaker_note %}} - Only the speaker can read these notes - Press `S` key to view {{% /speaker_note %}} </code></pre> <p>Press the <code>S</code> key to view the speaker notes!</p> <aside class="notes"> <ul> <li>Only the speaker can read these notes</li> <li>Press <code>S</code> key to view</li> </ul> </aside> <hr> <h2 id="themes">Themes</h2> <ul> <li>black: Black background, white text, blue links (default)</li> <li>white: White background, black text, blue links</li> <li>league: Gray background, white text, blue links</li> <li>beige: Beige background, dark text, brown links</li> <li>sky: Blue background, thin dark text, blue links</li> </ul> <hr> <ul> <li>night: Black background, thick white text, orange links</li> <li>serif: Cappuccino background, gray text, brown links</li> <li>simple: White background, black text, blue links</li> <li>solarized: Cream-colored background, dark green text, blue links</li> </ul> <hr> <section data-noprocess data-shortcode-slide data-background-image="/media/boards.jpg" > <h2 id="custom-slide">Custom Slide</h2> <p>Customize the slide style and background</p> <pre><code class="language-markdown">{{&lt; slide background-image=&quot;/media/boards.jpg&quot; &gt;}} {{&lt; slide background-color=&quot;#0000FF&quot; &gt;}} {{&lt; slide class=&quot;my-style&quot; &gt;}} </code></pre> <hr> <h2 id="custom-css-example">Custom CSS Example</h2> <p>Let&rsquo;s make headers navy colored.</p> <p>Create <code>assets/css/reveal_custom.css</code> with:</p> <pre><code class="language-css">.reveal section h1, .reveal section h2, .reveal section h3 { color: navy; } </code></pre> <hr> <h1 id="questions">Questions?</h1> <p><a href="https://github.com/wowchemy/wowchemy-hugo-modules/discussions" target="_blank" rel="noopener">Ask</a></p> <p><a href="https://wowchemy.com/docs/managing-content/#create-slides" target="_blank" rel="noopener">Documentation</a></p>