The Role of URLs in Objectionable Web Content Categorization

Over the past decade, significant advances have been made in the field of image search for e-commerce applications. Traditional image-to-image retrieval models, which focus solely on image details such as texture, tend to overlook useful semantic information contained within the images. As a result, the retrieved products might possess similar image details, but fail to fulfil the user's search goals. Moreover, the use of image-to-image retrieval models for products containing multiple images results in significant online product feature storage overhead and complex mapping implementations. In this paper, we report the design and deployment of the proposed Multi-modal Item Embedding Model (MIEM) to address these limitations. It is capable of utilizing both textual information and multiple images about a product to construct meaningful product features. By leveraging semantic information from images, MIEM effectively supplements the image search process, improving the overall accuracy of retrieval results. MIEM has become an integral part of the Shopee image search platform. Since its deployment in March 2023, it has achieved a remarkable 9.90% increase in terms of clicks per user and a 4.23% boost in terms of orders per user for the image search feature on the Shopee e-commerce platform.

Building machine learning (ML) tools, or systems, for use in manufacturing environments is a challenge that extends far beyond the understanding of the ML algorithm. Yet, these challenges, outside of the algorithm, are less discussed in literature. Therefore, the purpose of this work is to practically illustrate several best practices, and challenges, discovered while building an ML system to detect tool wear in metal CNC machining. Namely, one should focus on the data infrastructure first; begin modeling with simple models; be cognizant of data leakage; use open-source software; and leverage advances in computational power. The ML system developed in this work is built upon classical ML algorithms and is applied to a real-world manufacturing CNC dataset. The best-performing random forest model on the CNC dataset achieves a true positive rate (sensitivity) of 90.3% and a true negative rate (specificity) of 98.3%. The results are suitable for deployment in a production environment and demonstrate the practicality of the classical ML algorithms and techniques used. The system is also tested on the publicly available UC Berkeley milling dataset. All the code is available online so others can reproduce and learn from the results.

Local and global topics in text modeling of web pages nested in web sites

In recent days booking through web sites has become very prominent. Almost 70% of bookings or transactions are happening through online. Of all the sales web sites housing websites has become a great platform for sales. Different companies provide different services like selling new house, house rentals, sale of used houses etc. Price matters a lot in any sa les. As houses are one time buy in the life for most of the Indians, they do a lot of research related to price, area and facilities while buying a house. The first thing common middle-class person looks at is the price based on their requirement. Integration of Machine learning algorithm to these web sites or application can be a great advantage. This can reduce the manual effort of sending quotations or negotiations. Presently we see many web sites which give us the details regarding the houses and their prices, but a person must invest more time to search the house which fits his necessities and budget. By integrating a machine learning algorithm into these websites, it helps the user just to sort his necessities and saves his time. In this paper the usage of this algorithm is explained practically and how the functionality can help improve the deliverability of the application is also discussed.

Nowadays, in the network age, computer security issues have attracted much attention. In order to ensure computer network security, it is necessary to pay attention to improving data encryption technology. Data encryption technology is mainly divided into two types: asymmetric key and symmetric key. Generally, the sender and receiver use different password settings to realize network security. In this paper, a scheme of applying machine learning classification algorithm to homomorphic encrypted data sets is proposed: firstly, the plaintext is preprocessed to ensure that it meets the requirements of homomorphic encryption of data; Then, compare and sort the encrypted data set by protocol. Finally, the classification results are obtained. Combined with machine learning algorithm, the text information hiding convergence is controlled, and the text information hiding algorithm is optimized. Simulation results show that this method can hide text information in a higher depth, and has stronger anti-attack ability, thus improving the security of text information storage.

Googling for Health Information

The abstract describes the construction of a binary classification model for predicting the type of job advertisement in cloud-based ATS (Applicant Tracking Systems) as either legitimate or fraudulent. Various machine learning algorithms can be employed to address this issue. Traditional classification algorithms, including LSVC (Support Vector Machine), GBT (Gradient Boosting Tree), and RF (Random Forest), have been chosen for this study. One approach to building such a model involves identifying and collecting relevant attributes or features that can help distinguish fraudulent job advertisements from legitimate ones. Some features that could be useful in detecting fraudulent job ads include job location, job description, job requirements, job responsibilities, company information, and recruiter data. Subsequently, different machine learning algorithms can be trained on prepared datasets using standard methods such as cross-validation to assess their performance. The performance of the trained models can be evaluated using various metrics such as accuracy, precision, and recall. Ultimately, the most effective model can be selected based on these evaluation metrics and deployed in a production environment, where it can classify job advertisements as fraudulent or legitimate. It's important to note that the model should also undergo continuous evaluation and updates over time to ensure its reliability and effectiveness. Based on the evaluation metrics, it was concluded that the GBT classifier exhibits higher performance and accuracy compared to the LinearSVC and RF classifiers on the given dataset. However, it should be considered that the GBT classifier requires more time for training and prediction; GBT takes 208.738579 seconds, while LSVC and RF take 64.267132 and 71.024914 seconds, respectively. Taking into account the evaluation results, the GBT model was utilized for the operational aspect of the program. For implementation of the prediction, machine learning was performed on GBT, RF, and LSVC using a custom dataset called "Job_Fraud," created based on the publicly available EMSCAD dataset. To address the significant data imbalance, an implementation of the Synthetic Minority Over-sampling Technique (SMOTE) from a library was utilized. Initially, a model was obtained and trained on the data using a classifier, removing stop-words through TFIDFVectorizer in the vector space. Then, after reducing the dimensionality of the data, the data was reloaded, and both the model and vectorizer were retrained before being used for prediction. The tkinter module was used for the graphical interface. The predict() function utilizes the trained model for predictions based on the feature vector.

Due to the rapid growth of the number of digital images on the Web, there is an increasing demand for an effective and efficient method for organizing and retrieving the available images. This article describes iFind, a system for clustering and searching WWW images. By using a vision-based page segmentation algorithm, a Web page is partitioned into blocks, and the textual and link information of an image can be accurately extracted from the block containing that image. The textual information is used for image indexing. By extracting the page-to-block, block-to-image, block-to-page relationships through link structure and page layout analysis, we construct an image graph. Our method is less sensitive to noisy links than previous methods like PageRank, HITS, and PicASHOW, and hence the image graph can better reflect the semantic relationship between images. Using the notion of Markov Chain, we can compute the limiting probability distributions of the images, ImageRanks, which characterize the importance of the images. The ImageRanks are combined with the relevance scores to produce the final ranking for image search. With the graph models, we can also use techniques from spectral graph theory for image clustering and embedding, or 2-D visualization. Some experimental results on 11.6 million images downloaded from the Web are provided in the article.

Purpose – Project ATHENA aims to develop an application to address information overload, primarily focused on Recommendation Systems (RSs) with the personalization and user experience design of a modern system. Method – Two machine learning (ML) algorithms were used: (1) TF-IDF for Content-based filtering (CBF); (2) Classification with Matrix Factorization- Singular Value Decomposition (SVD) applied with Collaborative filtering (CF) and mean (normalization) for prediction accuracy of the CF. Data sampling in academic Research and Development (R&D) of Philippine Council for Agriculture, Aquatic, and Natural Resources Research and Development (PCAARRD) e-Library and Project SARAI publications plus simulated data used as training sets to generate a recommendation of items that uses the three RS filtering (CF, CBF, and personalized version of item recommendations). Series of Testing and TAM performed and discussed. Results – Findings allow users to engage in online information and quickly evaluate retrieved items produced by the application. Compatibility-testing (CoT) shows the application is compatible with all major browsers and mobile-friendly. Performance-testing (PT) recommended v-parameter specs and TAM evaluations results indicate strongly associated with overall positive feedback, thoroughly enough to address the information-overload problem as the core of the paper. Conclusion – A modular architecture presented addressing the information overload, primarily focused on RSs with the personalization and design of modern systems. Developers utilized Two ML algorithms and prototyped a simplified version of the architecture. Series of testing (CoT and PT) and evaluations with TAM were performed and discussed. Project ATHENA added a UX feature design of a modern system. Recommendations – High-end hardware specs of v-Parameter are recommended, at least 8-cores of vCPU with 16GiB of memory and sufficient driver-type size to run the model and End-to-end jobs execution for continuously maintaining the application. Research Implications – Future Developers must use/integrate a large dataset. Other ML approaches can expand Hybrid-RS better. Implicit data and additional filtering methods may enhance the application in the future.

This paper presents a fuzzy Web site evaluator. It extracts personality and emotion cues from textual analysis in a thesaurus. User-provided weights characterize semantic fuzzy grades in the thesaurus. Textual information is extracted from Web pages' titles, section headings and full-text contents, in order of decreasing importance. Colors of images in Web pages are also treated and related to both personality and emotion. The output consists of bar diagrams summarizing the personality and emotional content of the specified Web sites.

Currently, web image search is mostly based on textual information associated with Web pages. However, this search method ignores visual content information of images. Visual search reranking that aims to improve the text-based image search results by leveraging visual content analysis. This paper presents a semi-supervised learning to rank framework based on Ranking SVM. To better compute the similarity between images, a new similarity measure algorithm named SM-PCA, which is relying on Principle Component Analysis, is proposed and introduced into semi-supervised learning. Finally, the proposed method is evaluated by image database downloaded from the popular image search engine. Experiments show that our method outperforms the state-of-the-art method.KeywordsImage search rerankingSM-PCARanking SVM

A simple and robust wetland classification approach by using optical indices, unsupervised and supervised machine learning algorithms

Recent years have witnessed a great success of social media websites. Tag-based image search is an important approach to access the image content of interest on these websites. However, the existing ranking methods for tag-based image search frequently return results that are irrelevant or lacking in diversity. Most of the existing methods estimate the relevance of images by using tags and visual characteristics either separately or sequentially. The proposed system uses an approach that utilize simultaneously both visual information and textual information in real time to estimate the relevance of user tagged image. The method used to determine the relevance estimation is the hypergraph learning approach. The hypergraph is a generalization of a graph in which an edge in the hypergraph can be connected to any number of vertices. In the proposed method each social image can be represented by the bag-of-visual words and bag-of-textual words, which can be obtained from the textual content and visual content of the particular image. A hypergraph can be constructed in which the vertices represent the social images for ranking and the each hyperedge represents the visual words or tags that are obtained from the image. In the hypergraph learning scheme, both the visual content and tag information are taken into consideration at same time. Different from the method used by the traditional hypergraph, in the proposed system a social image hypergraph is constructed where vertices represent the images and hyperedges represent the visual or textual terms. The set of pseudo-positive images are used to achieve the learning, where the weight of hyperedges are updated throughout the learning process. Thus only the most relevant images are given to the user.

The amount of high-quality data in the Web databases has been increasing dramatically. To utilize such wealth of information, measuring the similarity betweenWeb databases has been proposed for many applications, such as clustering and top-k recommendation. Most of the existing methods use the text information either in the interfaces of Web databases or in the Web pages where the interfaces are located, to represent the Web databases. These methods have the limitation that the text may contain a lot of noisy words, which are rarely discriminative and cannot capture the characteristics of the Web databases. To better measure the similarity between Web databases, we introduce a novel Web database similarity method.We employ the categories of the records in the Web databases, which can be automatically extracted from the Web sites where the Web databases are located, to represent the Web databases. The record categories are of high-quality and can capture the characteristics of the corresponding Web databases effectively. In order to better utilize the record categories, we measure the similarity between Web databases based on a unified category hierarchy, and propose an effective method to construct the category hierarchy from the record categories obtained from all the Web databases. We conducted experiments on real ChineseWeb Databases to evaluate our method. The results show that our method is effective in clustering and top-k recommendation for Web Databases, compared with the baseline method, and can be used in real Web database related applications.KeywordsWeb databasesimilarity modelcategory hierarchy

The development of computational data science techniques in natural language processing (NLP) and machine learning (ML) algorithms to analyze large and complex textual information opens new avenues to study intricate policy processes at a scale unimaginable even a few years ago. We apply these scalable NLP and ML techniques to analyze the United States Government’s regulation of the banking and financial services sector. First, we employ NLP techniques to convert the text of financial regulation laws into feature vectors and infer representative “topics” across all the laws. Second, we apply ML algorithms to the feature vectors to predict various attributes of each law, focusing on the amount of authority delegated to regulators. Lastly, we compare the power of alternative models in predicting regulators’ discretion to oversee financial markets. These methods allow us to efficiently process large amounts of documents and represent the text of the laws in feature vectors, taking into account words, phrases, syntax, and semantics. The vectors can be paired with predefined policy features, thereby enabling us to build better predictive measures of financial sector regulation. The analysis offers policymakers and the business community alike a tool to automatically score policy features of financial regulation laws to and measure their impact on market performance.