Intelligent Question Answering Module for Product Manuals

Question Answering (QA) has been a well-researched NLP problem over the past few years. The ability for users to query through information content that is available in a range of formats - organized and unstructured - has become a requirement. This paper proposes to untangle factoid question answering targeting the Hi-Tech domain. This paper addresses issues faced during document question answering, such as document parsing, indexing and retrieval (identifying the relevant documents) as well as machine comprehension (extract spans of correct answers from the context). Our suggested solution provides a comprehensive pipeline comprised of document ingestion modules that handle a wide range of unstructured data across various sections of the document, such as textual, images, and tabular content. Our studies on a variety of “real-world” and domain-specific datasets show how current fine-tuned models are insufficient for this challenging task, and how our proposed pipeline is an effective alternative.

Fast and Effective Retrieval for Large Multimedia Collections

The indexing and retrieval of multimedia content is generally implemented by employing feature graphs. These graphs typically contain a significant number of nodes and edges to reflect the level of detail in feature detection. A higher level of detail increases the effectiveness of the results, but also leads to more complex graph structures. However, graph traversal-based algorithms for similarity are quite inefficient and computationally expensive, especially for large data structures. To deliver fast and effective retrieval especially for large multimedia collections and multimedia big data, an efficient similarity algorithm for large graphs in particular is desirable. Hence, in this paper, we define a graph projection into a 2D space (Graph Code) and the corresponding algorithms for indexing and retrieval. We show that calculations in this space can be performed more efficiently than graph traversals due to the simpler processing model and the high level of parallelization. As a consequence, we demonstrate experimentally that the effectiveness of retrieval also increases substantially, as the Graph Code facilitates more levels of detail in feature fusion. These levels of detail also support an increased trust prediction, particularly for fused social media content. In our mathematical model, we define a metric triple for the Graph Code, which also enhances the ranked result representations. Thus, Graph Codes provide a significant increase in efficiency and effectiveness, especially for multimedia indexing and retrieval, and can be applied to images, videos, text and social media information.

A distributed Content-Based Video Retrieval system for large datasets

With the rapid growth in the amount of video data, efficient video indexing and retrieval methods have become one of the most critical challenges in multimedia management. For this purpose, Content-Based Video Retrieval (CBVR) is nowadays an active area of research. In this article, a CBVR system providing similar videos from a large multimedia dataset based on query video has been proposed. This approach uses vector motion-based signatures to describe the visual content and uses machine learning techniques to extract key frames for rapid browsing and efficient video indexing. The proposed method has been implemented on both single machine and real-time distributed cluster to evaluate the real-time performance aspect, especially when the number and size of videos are large. Experiments were performed using various benchmark action and activity recognition datasets and the results reveal the effectiveness of the proposed method in both accuracy and processing time compared to previous studies.

Efficient indexing and retrieval of patient information from the big data using MapReduce framework and optimisation

Large and complex data becomes a valuable resource in biomedical discovery, which is highly facilitated to increase the scientific resources for retrieving the helpful information. However, indexing and retrieving the patient information from the disparate source of big data is challenging in biomedical research. Indexing and retrieving the patient information from big data is performed using the MapReduce framework. In this research, the indexing and retrieval of information are performed using the proposed Jaya-Sine Cosine Algorithm (Jaya–SCA)-based MapReduce framework. Initially, the input big data is forwarded to the mapper randomly. The average of each mapper data is calculated, and these data are forwarded to the reducer, where the representative data are stored. For each user query, the input query is matched with the reducer, and thereby, it switches over to the mapper for retrieving the matched best result. The bilevel matching is performed while retrieving the data from the mapper based on the distance between the query. The similarity measure is computed based on the parametric-enabled similarity measure (PESM), cosine similarity and the proposed Jaya–SCA, which is the integration of the Jaya algorithm and the SCA. Moreover, the proposed Jaya–SCA algorithm attained the maximum value of F-measure, recall and precision of 0.5323, 0.4400 and 0.6867, respectively, using the StatLog Heart Disease dataset.

AksharaNet: A GPU Accelerated Modified Depth-Wise Separable Convolution for Kannada Text Classification

For content-based indexing and retrieval applications, text characters embedded in images are a rich source of information. Owing to their different shapes, grayscale values, and dynamic backgrounds, these text characters in scene images are difficult to detect and classify. The complexity increases when the text involved is a vernacular language like Kannada. Despite advances in deep learning neural networks (DLNN), there is a dearth of fast and effective models to classify scene text images and the availability of a large-scale Kannada scene character dataset to train them. In this paper, two key contributions are proposed, AksharaNet, a graphical processing unit (GPU) accelerated modified convolution neural network architecture consisting of linearly inverted depth-wise separable convolutions and a Kannada Scene Individual Character (KSIC) dataset which is grounds-up curated consisting of 46,800 images. From results it is observed AksharaNet outperforms four other well-established models by 1.5% on CPU and 1.9% on GPU. The result can be directly attributed to the quality of the developed KSIC dataset. Early stopping decisions at 25% and 50% epoch with good and bad accuracies for complex and light models are discussed. Also, useful findings concerning learning rate drop factor and its ideal application period for application are enumerated.