Information Retrieval Process Research Articles

Content Summarization and Question Answering System Using LLM

This paper presents a novel approach to content summarization, a pivotal task in information processing and knowledge extraction. Our methodology integrates cutting-edge technologies, including large language models (LLMs) and ad-vanced retrieval techniques, to efficiently distill extensive textual data into concise and informative summaries. By leveraging Anyscale LLMs for language comprehension and employing prompt engineering for model guidance, we propose a frame-work that prioritizes efficient document indexing and rapid retrieval using DuckDB vector stores. Additionally, we introduce the RetrievalQAWithSourcesChain framework, which combines LLMs with retrievers for proficient question answering. Em-pirical validation demonstrates the efficacy of our approach in generating high-quality content summaries with minimal manual intervention, thereby enhancing information retrieval and knowledge extraction processes. Index Terms—LLMs, Anyscale, prompt engineering, document indexing, DuckDB, vector stores, RetrievalQAWithSourcesChain framework, question answering, empirical validation

A decentralized solid compound storage facility managed by a centralized electronic platform at a growing drug discovery company

Within a growing drug discovery company, scientists acquire (either through in house synthesis or purchase) then store, retrieve, and ship solid compound samples daily between multiple locations. The efficient management and tracking of this entire process to support drug discovery is a significant challenge. This article describes a decentralized and cost-effective inventory facility that simplifies the solid compound storage and retrieval process. Standardized storage cabinets from the market are utilized, providing a cost-effective physical infrastructure. The cabinets can be distributed across storage rooms at multiple sites and arranged into spaces with a variety of dimensions, allowing the system to be retrofitted into existing facilities and scaled up easily. We can provide storage close to work areas at each location, minimizing both unnecessary movement of staff and transportation of substances. We have applied a systematic barcoding method to the compound batch identifier that correlates with its compound location. This simplifies the compound registration process as well as the process of finding and returning compounds. Additionally, a centralized electronic platform has been employed to store, update and track solid compound information, such as properties, location and quantity. Compound shipment may be initiated from different sites, and a centralized electronic platform assists the information retrieval process, ensuring each location possesses up-to-date information. The electronic platform we present streamlines the management of compound registration, location tracking, weight updates and shipment information, facilitating seamless record sharing among all stakeholders. Every step of the process can be tracked in real time by the project team. The platform can be flexibly configured to adapt to an evolving set of storage locations, with all information and processes being audited.

ADVANCMENTS IN TEXT SUMMARIZATION AND EXTRACTIVE QUESTION- ANSWERING : A MACHINE LEARNING APPROACH

In the era of social media platforms, the rapid expansion of data mining in the fields of information retrieval and natural language processing emphasizes the crucial need for automated text summarization. At the current time, pretrained word embedding techniques and sequence to sequence models can be effectively repurposed in the realm of social network summarization to efficiently condense significant information with strong encoding capabilities. However, dealing with the challenge of extended text dependency and efficiently utilizing latent topic mapping presents an increasingly significant obstacle for these computational models. In this document, we propose a topic- focused approach for both extractive and abstractive summarization, integrating Question Answering features by employing BERT and Pegasus pretrained models. A comprehensive analysis of the architectural complexities of the pre-existing models utilized in the creation of our model has been presented. The evaluation of our Text Summarization & Question Answering Model was conducted meticulously on a variety of datasets including Multi news, XSum, and CNN/Daily Mail, with experimental results indicating its achievement of state-of-the-art performance levels based on ROGUE scores. Following this, the verification of our results through human evaluation confirms that the summaries generated by our model are in alignment with the performance standards set by humans across various datasets. Furthermore, we have integrated an Extractive Question Answering task into our model, proposing an architecture leveraging BERT and conducting comparative analyses against alternative language models to gauge its efficacy. Keywords—Natural Language Processing (NLP), Text Summarization, Question & Answering (Q&A), Transformer Models, Pegasus, BERT (Bidirectional Encoder Representations from Transformers), Neural Topic Modeling, Extractive Summarization, Abstractive Summarization, Semantic Representation, Topic Embedding

O conceito de relevância na recuperação de informação

This work aims to present a bibliographical survey on the concept of relevance, its properties and characteristics, as well as systematizing its manifestations and its use in Information Science, particularly in the process of information retrieval. To achieve this, the research was based on a bibliographical survey of several authors who have relevance as an object of study. It found that the concept of relevance has several formal definitions and can be interpreted in different ways. Relevance is also strongly linked to information retrieval systems. In this context, it can be classified under two approaches: user relevance and system relevance. From the results, we conclude that relevance plays a fundamental role in Information Science, with emphasis on the information retrieval process, which often uses this concept as a metric to evaluate the effectiveness of algorithms and systems.

A novel fuzzy logic-based approach for textual documents indexing

In the evolving landscape of information retrieval and natural language processing, the quest for more effective automatic keyword extraction (AKE) techniques from textual documents has become a pivotal research focus. Existing methodologies, while offering valuable insights, often grapple with the challenges posed by the imprecision and variability inherent in human language. This has led to a growing recognition of the need for innovative approaches to navigating textual content’s nuances more adeptly. In response to this imperative, this paper proposes a novel fuzzy indexing approach designed specifically for the indexing of textual documents. Fuzzy indexing, grounded in the principles of fuzzy logic, provides solutions for handling the inherent uncertainty and imprecision in natural language, especially when confronted with the intricacies of linguistic ambiguity and variability. By leveraging the power of fuzzy logic, we aim to enhance the precision of keyword extraction. This paper unfolds the intricacies of our fuzzy indexing approach, detailing the theoretical methodology through empirical evaluation and comparative analysis; we seek to demonstrate the efficacy of our approach in outperforming traditional methods in the context of fuzzy indexing for textual documents.

From traditional to tech-savvy: the evolution of Nigerian libraries in the machine learning era

Purpose This study explores the implementation of personalised information access, driven by machine learning, in Nigerian public libraries. The purpose of this paper is to address existing challenges, enhance the user experience and bridge the digital divide by leveraging advanced technologies. Design/methodology/approach This study assesses the current state of Nigerian public libraries, emphasising challenges such as underfunding and lack of technology adoption. It proposes the integration of machine learning to provide personalised recommendations, predictive analytics for collection development and improved information retrieval processes. Findings The findings underscore the transformative potential of machine learning in Nigerian public libraries, offering tailored services, optimising resource allocation and fostering inclusivity. Challenges, including financial constraints and ethical considerations, are acknowledged. Originality/value This study contributes to the literature by outlining strategies for responsible implementation and emphasising transparency, user consent and diversity. The research highlights future directions, anticipating advancements in recommendation systems and collaborative efforts for impactful solutions.

The effect of augmented reality applied to learning process with different learning styles in structural engineering education

PurposeThe purpose of this study is to scrutinize the influence of individual learning styles on the effectiveness of augmented reality (AR)-based learning in structural engineering. There has been a lack of research examining the correlation between learning efficiency and learning style, particularly in the context of quantitatively assessing the efficacy of AR in structural engineering education.Design/methodology/approachUsing Kolb’s experiential learning theory (ELT), a model that emphasizes learning through experience, students from the construction management department are assigned four learning styles (converging, assimilating, diverging and accommodating). Performance data were gathered, appraised, and compared through the three dimensions from the Knowledge, Attitude and Practices (KAP) survey model across four categories of Kolb’s learning styles in both text-graph (TG)-based and AR-based learning settings.FindingsThe findings indicate that AR-based materials positively impact structural engineering education by enhancing overall learning performance more than TG-based materials. It is also found that the learning style has a profound influence on learning effectiveness, with AR technology markedly improving the information retrieval processes, particularly for converging and assimilating learners, then diverging learners, with a less significant impact on accommodating learners.Originality/valueThese results corroborate prior research analyzing learners' outcomes with hypermedia and informational learning systems. It was found that learners with an “abstract” approach (convergers and assimilators) outperform those with a “concrete” approach (divergers and accommodators). This research emphasizes the importance of considering learning styles before integrating technologies into civil engineering education, thereby assisting software developers and educational institutions in creating more effective teaching materials tailored to specific learning styles.

Leveraging Large Language Models for Tradespace Exploration

This paper proposes a method for leveraging large language models (LLMs) to improve the question-answering capabilities of artificial intelligence (AI) assistants for tradespace exploration. The method operates by querying an information space composed of fused data sources encompassing the tradespace exploration process and responding based on the gathered information. The information retrieval process is modeled as an internal dialog where an LLM-based dialog agent converses with a subquery answering agent. A case study is conducted on a next-generation soil moisture mission (SM-NG), and a generative AI assistant (named Daphne-G) is configured on it. The effect of the dialog agent and the choice of LLM are assessed by comparing the performance of three different system configurations on a validation question set. A second validation effort is conducted, comparing Daphne-G’s responses to those of a baseline template-based AI assistant, Daphne-VA. Results show that the dialog-based system is necessary for answering complex questions requiring multiple documents. Furthermore, results show that Daphne-G can correctly answer all the questions Daphne-VA can answer, while simultaneously being able to answer a greater number of questions than Daphne-VA. The results suggest that LLMs could significantly improve the outcomes of the tradespace exploration process, which may result in better and more cost-effective mission concepts being implemented.

Radiation injuries of organs and tissues: mechanisms of occurrence, methods of prevention and treatment: A review

In modern medicine, radiation therapy (RT) is one of the main therapies for cancer. According to the World Health Organization, about 70% of patients diagnosed with cancer require RT at some stage of treatment. Development and improvement of methods for early diagnosis of cancers, multimodal therapy for cancer patients, including surgical intervention, new drug treatment regimens, modern methods of conformal 3D-RT and proton therapy significantly improved the treatment outcomes of patients and increased their life expectancy. Modern RT is a high-tech chain of sequential use of complex technical devices and mathematical calculations to destroy the tumor without adversely affecting the surrounding normal tissues. Some patients develop radiation injuries to normal organs and tissues that reduce their quality of life. Prevention and treatment of such injuries are urgent problems of modern oncology. The article addresses the pathogenesis and mechanism of the development of systemic and local radiation responses, as well as methods of treatment and prevention of some of them. Three electronic databases, including PubMed, Cochrane and EMBASE, and Russian radiotherapists' publications, were used for systematic review in the information retrieval process.

A Novel Approach to Text Summarization Using Machine Learning

Text summarization is a key strategy in the domains of information retrieval and natural language processing (NLP). Its objective is to reduce a lengthy written document into a clearer, more succinct summary of the information it contains. When a text document is too lengthy or intricate to analyse completely, as in news stories, academic papers, or legal documents, this approach is extremely helpful. The major challenge of text summarising is to take the most important and relevant information from the original text and convey it in an understandable and concise way. In this study, extractive and abstractive summarising techniques are the two primary categories of text summary methods. The paper also presents several algorithms that have been proposed for text summarization, including TextRank, Seq2Seq, and BART. TextRank is a simple and fast algorithm that works well for short documents, Seq2Seq is a deep learning-based approach that generates high-quality summaries, and BART is a transformer-based algorithm that provides the best results on benchmark datasets. The obtained ROUGE Score after passing TextRank, BART, and Seq2SEq algorithm significant also.

Assessment of Information Literacy Competences in Peruvian University Students: An Exploratory Study

The objective was to evaluate the perception of the informational literacy competences of the university students of the Peruvian Amazon. The investigation was developed under a quantitative approach, the design was non-experimental and cross-sectional descriptive type. The sample was made up of 707 students to whom the Informational Literacy Competences Questionnaire was given, an instrument with an adequate level of content validity and reliability. According to the results, the level of development of the informational literacy competencies of 37.5% of students was low, 35% was medium, and 27.5% was high. Likewise, the dimensions of information retrieval, information assessment and information processing were also valued at a low level. However, the information communication dimension of information was valued at a medium level. On the other hand, it was found that the level of informational literacy competences was significantly associated with the year of study in which the students were (p<0.05). It was concluded that the level of development of the informational literacy competencies that characterized the students was low.

Neurophysiological processes reflecting the effects of the immediate past during the dynamic management of actions

In recent years, there has been many efforts to establish a comprehensive theoretical framework explaining the working mechanisms involved in perception-action integration. This framework stresses the importance of the immediate past on mechanisms supporting perception-action integration. The present study investigates the neurophysiological principles of dynamic perception-action bindings, particularly considering the influence of the immediate history on action control mechanisms. For this purpose, we conducted an established stimulus-response binding paradigm during EEG recording. The SR-task measures stimulus-response binding in terms of accuracy and reaction time differences depending on the degree of feature overlap between conditions. Alpha, beta and theta band activity in distinct time domains as well as associated brain regions were investigated applying time-frequency analyses, a beamforming approach as well as correlation analyses. We demonstrate, for the first time, interdependencies of neuronal processes relying on the immediate past. The reconfiguration of an action seems to overwrite immediately preceding processes. The analyses revealed modulations of theta (TBA), alpha (ABA) and beta band activity (BBA) in connection with fronto-temporal structures supporting the theoretical assumptions of the considered conceptual framework. The close interplay of attentional modulation by gating irrelevant information (ABA) and binding and retrieval processes (TBA) is reflected by the correlation of ABA in all pre-probe-intervals with post-probe TBA. Likewise, the role of BBA in maintaining the event file until retrieval is corroborated by BBA preceding the TBA-associated retrieval of perception-action codes. Following action execution, TBA shifted towards visual association cortices probably reflecting preparation for upcoming information, while ABA and BBA continue to reflect processes of attentional control and information selection for goal-directed behavior. The present work provides the first empirical support for concepts about the neurophysiological mechanisms of dynamic management of perception and action.

Improved Object Detection with Content and Position Separation in Transformer

In object detection, Transformer-based models such as DETR have exhibited state-of-the-art performance, capitalizing on the attention mechanism to handle spatial relations and feature dependencies. One inherent challenge these models face is the intertwined handling of content and positional data within their attention spans, potentially blurring the specificity of the information retrieval process. We consider object detection as a comprehensive task, and simultaneously merging content and positional information like before can exacerbate task complexity. This paper presents the Multi-Task Fusion Detector (MTFD), a novel architecture that innovatively dissects the detection process into distinct tasks, addressing content and position through separate decoders. By utilizing assumed fake queries, the MTFD framework enables each decoder to operate under a presumption of known ancillary information, ensuring more specific and enriched interactions with the feature map. Experimental results affirm that this methodical separation followed by a deliberate fusion not only simplifies the task difficulty of the detection process but also augments accuracy and clarifies the details of each component, providing a fresh perspective on object detection in Transformer-based architectures.

Toward a Model to Evaluate Machine-Processing Quality in Scientific Documentation and Its Impact on Information Retrieval

The lack of quality in scientific documents affects how documents can be retrieved depending on a user query. Existing search tools for scientific documentation usually retrieve a vast number of documents, of which only a small fraction proves relevant to the user’s query. However, these documents do not always appear at the top of the retrieval process output. This is mainly due to the substantial volume of continuously generated information, which complicates the search and access not properly considering all metadata and content. Regarding document content, the way in which the author structures it and the way the user formulates the query can lead to linguistic differences, potentially resulting in issues of ambiguity between the vocabulary employed by authors and users. In this context, our research aims to address the challenge of evaluating the machine-processing quality of scientific documentation and measure its influence on the processes of indexing and information retrieval. To achieve this objective, we propose a set of indicators and metrics for the construction of the evaluation model. This set of quality indicators have been grouped into three main areas based on the principles of Open Science: accessibility, content, and reproducibility. In this sense, quality is defined as the value that determines whether a document meets the requirements to be retrieved successfully. To prioritize the different indicators, a hierarchical analysis process (AHP) has been carried out with the participation of three referees, obtaining as a result a set of nine weighted indicators. Furthermore, a method to implement the quality model has been designed to support the automatic evaluation of quality and perform the indexing and retrieval process. The impact of quality in the retrieval process has been validated through a case study comprising 120 scientific documents from the field of the computer science discipline and 25 queries, obtaining as a result 21% high, 39% low, and 40% moderate quality.

Report on the 13th Italian Information Retrieval Workshop (IIR 2023)

The 13th Italian Information Retrieval Workshop is the thirteenth edition of the annual conference of the Italian information retrieval and recommender systems communities. The two days of the conference gathered interesting studies and research work on a wide range of topics on information retrieval, recommender systems, and natural language processing, such as Search and Ranking, Recommendation, Content Analysis, and Classification, Artificial Intelligence, NLP, Semantics, and Dialog, Domain-Specific Applications, Human Factors and Interfaces, and Evaluation. It was hosted by the National Research Council (CNR) of Italy and the University of Pisa in a conference facility in Pisa, Italy. Date : 8--9 June 2023. Website : http://iir2023.isti.cnr.it/.

Large Language Models and Information Retrieval

This research article explores the synergistic integration of Optical Character Recognition (OCR) technology and Large Language Models (LLMs) to advance Information Retrieval (IR) processes. In a data-centric society, efficient IR is imperative, and the combination of OCR and LLMs presents a powerful solution. OCR transforms diverse document types into machine-readable formats, while LLMs excel in language understanding and generation. The article delves into the technical intricacies of these technologies, their seamless integration, and their potential to revolutionize information retrieval. By investigating their collaborative capabilities, this research contributes to the evolving landscape of natural language processing and information retrieval systems.

Transformasi Tata Kelola Arsip di Kantor Wali Nagari Dilam, Kec. Bukit Sundi, Kab. Solok

Rapid technological developments are not directly proportional to the management of records in an agency. This is evidenced by the fact that there are still many agencies that still manage their archives manually so that it takes a long time for the processing to reach the information retrieval process. One agency that has not yet managed its archives electronically is the Wali Nagari Dilam Office, Solok Regency. Therefore, there is a need for training related to managing records electronically using the ARTERI application to make it easier for village officials to manage their records. After conducting the training, there was an increase in the ability of the participants by as much as 70% so it is hoped that it will increase again so that all archives can be managed properly and can be found quickly and precisely.

Toward an Educative EEG-Based neuroIIR System for Adapting Contents

The pursuit of information is an integral aspect of a learner’s daily routine, often facilitated by information retrieval systems. In this study, we explore the correlation between text complexity and reading comprehension through the lens of neuroinformatics. Our objective is to ascertain whether it is feasible to infer the difficulty level of a text for a learner based on their brain activity, quantified by an electroencephalogram (EEG), during the information retrieval process. To accomplish this, we administered a section of the Woodcock Reading Mastery Test to our participants, evaluating their average reading fluency. Subsequently, the 18 participants under observation read paragraphs of varying complexity and responded to questions pertaining to the text. Utilizing the amassed data, we trained a deep learning model known as EEGNet to autonomously discern the complexity of the text being read, relying on the EEG signals. Furthermore, we conducted a comparison with previous research that employed a distinct characterization approach and employed a random forest algorithm for classification. The outcomes revealed that the present approach (EEGNet) outperformed the former method, achieving an accuracy rate of 80.83% across all subjects, whereas the random forest algorithm achieved a mere 50.28% accuracy. Our findings suggest the potential to classify the EEG signals in accordance with the level of difficulty a student encounters when comprehending the text. We propose several avenues for extending our work into real-time scenarios.

Efficient information retrieval model: overcoming challenges in search engines-an overview

<span>Search engines play a vital role in information retrieval (IR) indexing and processing vast and diverse data, which now encompasses the ever-expanding wealth of multimedia content. However, search engine performance relies on the efficiency and effectiveness of their information retrieval systems (IRS). To enhance search engine performance, there is a need to develop more efficient and accurate IRS that retrieves relevant information quickly and accurately. To address this challenge, various approaches, including inverted indexing, query expansion, and relevance feedback, have been proposed for IR. Although these approaches have shown promising results, but their effectiveness and limitations require a comprehensive examination This research aims to investigate the challenges and opportunities in designing an efficient IRS for search engines and identify key areas for improvement and future research. The study involves a comprehensive literature review on information retrieval impacting academia, industry, healthcare, e-commerce, and other domains. Researchers rely on search engines to access relevant scientific papers, professionals use them to gather market intelligence, and consumers utilize them for product research and decision-making. The findings of this study will contribute to the development of more efficient and effective information retrieval systems, leading to improved search engine performance and user satisfaction.</span>

Leveraging Generative AI and Large Language Models: A Comprehensive Roadmap for Healthcare Integration.

Generative artificial intelligence (AI) and large language models (LLMs), exemplified by ChatGPT, are promising for revolutionizing data and information management in healthcare and medicine. However, there is scant literature guiding their integration for non-AI professionals. This study conducts a scoping literature review to address the critical need for guidance on integrating generative AI and LLMs into healthcare and medical practices. It elucidates the distinct mechanisms underpinning these technologies, such as Reinforcement Learning from Human Feedback (RLFH), including few-shot learning and chain-of-thought reasoning, which differentiates them from traditional, rule-based AI systems. It requires an inclusive, collaborative co-design process that engages all pertinent stakeholders, including clinicians and consumers, to achieve these benefits. Although global research is examining both opportunities and challenges, including ethical and legal dimensions, LLMs offer promising advancements in healthcare by enhancing data management, information retrieval, and decision-making processes. Continued innovation in data acquisition, model fine-tuning, prompt strategy development, evaluation, and system implementation is imperative for realizing the full potential of these technologies. Organizations should proactively engage with these technologies to improve healthcare quality, safety, and efficiency, adhering to ethical and legal guidelines for responsible application.