Forecasting accuracy is bounded by the information available about the future. This paper makes that statement precise using information-theoretic tools. Under logarithmic loss, the expected performance of any probabilistic forecast decomposes into two parts: an irreducible component and an approximation component. The irreducible term is the conditional entropy of the future given the available information, while the approximation term is the divergence between the true conditional distribution and the forecasting method. The gap between this conditional-entropy limit and an unconditional baseline is exactly the mutual information between the future observation and the declared information set. This leads to a definition of forecastability as the maximum achievable reduction in expected log loss. Evaluated across horizons, forecastability forms a profile that describes how predictive information varies with lead time. This profile reflects the dependence structure of the process and need not be monotone: predictive information may be concentrated at particular lags, including seasonal horizons, even when intermediate horizons contain little useful signal. From this profile, the paper defines the informative horizon set: the horizons at which forecastability exceeds a practical threshold. At horizons not in this set, the achievable gain over the unconditional baseline is necessarily small, regardless of the forecasting method used. The framework therefore separates what is learnable from what is not, and distinguishes limits imposed by the data from errors introduced by modelling. The result is a pre-modelling diagnostic that identifies where meaningful prediction is feasible before any model is chosen, providing a principled basis for allocating modelling effort across forecast horizons.
Wanjala Gidraf Joseph, Irene Wanjiru Muriithi, Mundu Mustafa Muhamad
et al.
The research explored the impact of advancing education opportunities in private universities on diversity and social inclusion in public universities in western Uganda. Institutional logic theory underpinned social inclusion and diversity in public universities, whereas marketization, neoliberalism, and institutional theories anchored opportunity expansion. The study adopted a cross-sectional design and quantitative approach. Seven hundred and forty-eight (748) lecturers constituted the target population, from which a sample of 302 responders was obtained. Information was collected through the Questionnaire. Data was analysed using Pearson r correlation coefficient at an alpha level of .05. Expanding opportunities and promoting social inclusion and diversity were found to be positively correlated (r (278) =.54, p =.030). The study concluded that social inclusion and diversity in public universities are positively correlated with the expansion of educational opportunities in private universities. The study recommended that, government implements policies that promote diversity and inclusion and prioritise funding for public universities.
[Purpose/Significance] In the era of mobile Internet, mobile phones are the most important information access tools. From the perspective of mobile phone use, this paper examines and reveals the information acquisition behavior of farmers, explores the information gap in rural areas, analyzes the characteristics of the information acquisition behavior of key groups, and provide insights into how to improve the information acquisition behavior and narrowing the information gap. The significance of the research is that, with a focus on digital inclusion, we start from the inclusive feature of the Internet and modern information technology, find out the key groups to pay attention to in the information gap, and put forward ideas on how to realize digital inclusion from the perspective of farmers' information acquisition and needs. [Method/Process] Based on the rural sample survey at the national level, principal component analysis and RIF regression analysis were used to measure farmers' information acquisition behavior and identify the role of key influencing factors in improving information acquisition behavior. This study analyzes farmers' information acquisition behavior from the perspective of mobile phone use, describes the micro characteristics of the information gap in rural areas, and makes the conclusion more scientific and generalizable based on the sample survey of farmers at the national level. [Results/Conclusions] The low "long tail" group in the evaluation results of farmers' information acquisition behavior exists in the aged people groups and the groups with the education level below junior middle school, which highlights the phenomenon of information gap in rural areas. Key explanatory variables have a more pronounced marginal effect on groups with low evaluation results of information acquisition behavior (below the median). Improving the impact of mobile phone use among key groups such as the elderly and the undereducated makes an important contribution to narrowing the information gap, which reflects the inevitable requirement of the concept of digital inclusion. Based on the improvement of network facilities, it is also necessary to pay attention to the use of mobile phone functions and information content acquisition of key groups, strengthen the accuracy of information supply, activate the information needs of key groups, and improve the adaptability of such groups in the digital environment. There are still more relatively poor and aging groups in rural areas, which need to be included in the digital inclusion path, improve the situation of vulnerable groups in the digital age, and provide the means to realize people's needs. Future research will pay more attention to promoting the provision of high-quality public service resources through digital tools, realizing the continuous empowerment of digital technology for rural development, and improving the digital literacy of citizens, especially rural residents.
Bibliography. Library science. Information resources, Agriculture
[Purpose/Significance] To explore in depth the acceptance and usage habits of AIGC tools by graduate students in the process of academic research, and to promote the positive attention and application of emerging technologies by graduate students is one of the goals of library knowledge service and information literacy education. This paper aims to reveal the influence mechanism of internal and external factors on the use of AIGC tools by graduate students at the user level, clarify the behavioral motivation of graduate students to use AIGC tools to support learning and research, help libraries to design and promote AIGC services according to the actual situation, and promote the implementation of AIGC technology in knowledge services. [Method/Process] Based on the UTAUT2 model, considering related theories such as perceived value and the characteristics of AIGC tool and graduate student group, this study constructed the influencing factor model of graduate students' AIGC tool use behavior, and provided empirical evidence through questionnaire survey and structural equation model analysis. The survey respondents are graduate students in universities or research institutes. In this study, questionnaires were distributed to graduate students through social media platforms, enterprise Wechat contacts, email, etc., and the survey period was from July to August 2024. After the data collection, statistical software such as SPSS and SmartPLS was used to analyze all the valid data obtained, including descriptive statistics, reliability and validity test and structural equation model analysis. [Results/ [Conclusions] Functional value, use value and emotional value in the tool aspect, individual innovation in individual aspect and social influence in environmental aspect have significant positive effects on graduate students' willingness to use AIGC tools, and indirectly affect their use behavior. Facilitating conditions, such as network equipment, as supporting factors, also have a significant positive impact on graduate students' usage. It is suggested that AIGC tool developers and library service designers consider the functional advantages and convenience. On the one hand, it is suggested that they pay attention to the functional value of the tool, that is, the auxiliary role to the graduate study and scientific research; on the other hand, they consider whether the tool is design-friendly, easy to operate, with low technical threshold and easy to use on an ongoing basis. From a graduate education perspective, it is important to promote the deep integration of the tool use with one's own professional learning and research in order to realize the improvement of other qualities through information literacy. Meanwhile, strengthening students' innovative thinking and comprehensive ability training, and guiding AIGC tool application ability and scientific research thinking to promote each other are conducive to new technologies to truly support learning and scientific research, and ultimately achieve the goal of developing high-level innovative talents.
Bibliography. Library science. Information resources, Agriculture
Marwah Alaofi, Negar Arabzadeh, Charles L. A. Clarke
et al.
In this chapter, we consider generative information retrieval evaluation from two distinct but interrelated perspectives. First, large language models (LLMs) themselves are rapidly becoming tools for evaluation, with current research indicating that LLMs may be superior to crowdsource workers and other paid assessors on basic relevance judgement tasks. We review past and ongoing related research, including speculation on the future of shared task initiatives, such as TREC, and a discussion on the continuing need for human assessments. Second, we consider the evaluation of emerging LLM-based generative information retrieval (GenIR) systems, including retrieval augmented generation (RAG) systems. We consider approaches that focus both on the end-to-end evaluation of GenIR systems and on the evaluation of a retrieval component as an element in a RAG system. Going forward, we expect the evaluation of GenIR systems to be at least partially based on LLM-based assessment, creating an apparent circularity, with a system seemingly evaluating its own output. We resolve this apparent circularity in two ways: 1) by viewing LLM-based assessment as a form of "slow search", where a slower IR system is used for evaluation and training of a faster production IR system; and 2) by recognizing a continuing need to ground evaluation in human assessment, even if the characteristics of that human assessment must change.
This paper introduces a novel information-theoretic perspective on the relationship between prominent group fairness notions in machine learning, namely statistical parity, equalized odds, and predictive parity. It is well known that simultaneous satisfiability of these three fairness notions is usually impossible, motivating practitioners to resort to approximate fairness solutions rather than stringent satisfiability of these definitions. However, a comprehensive analysis of their interrelations, particularly when they are not exactly satisfied, remains largely unexplored. Our main contribution lies in elucidating an exact relationship between these three measures of (un)fairness by leveraging a body of work in information theory called partial information decomposition (PID). In this work, we leverage PID to identify the granular regions where these three measures of (un)fairness overlap and where they disagree with each other leading to potential tradeoffs. We also include numerical simulations to complement our results.
Fatemeh Bahador, Azam Sabahi, Samaneh Jalali
et al.
Background and Aim: Diabetes is one of the most common metabolic diseases in Iran and the fifth leading cause of death all over the world. Its spread around the world has created new methods in biomedical research, including artificial intelligence. The present study was carried out to review the studies conducted in the area of artificial intelligence and diabetes in Iran.
Materials and Methods: This study was carried out using a systematic review method. Valid domestic databases, including Irandoc, Magiran, Sid and Google Scholar search engine, were reviewed using the keywords of artificial intelligence and diabetes in Persian both individually and in a combined manner without time limitation until June 20, 2021. A total number of 7495 articles were retrieved, which were screened in different stages (exclusion of duplicates (1824), title and summary of the articles (5884) and full text (30) and finally 20 articles that met the criteria desired by the researchers were carefully reviewed.
Results: Among the retrieved articles, 20 articles met the inclusion criteria, of which 16 articles dealt with methods based on artificial intelligence and 4 articles dealt with the design of new systems based on artificial intelligence. Also, 10 articles examined the role of artificial intelligence in prediction, 8 articles in diagnosis, and 2 articles dealt with the control and management of diabetes. Most of the articles were related to the use of data mining methods such as artificial neural network, decision tree, etc. (16 articles). Some studies also evaluated and compared artificial intelligence methods on application, accuracy and the sensitivity of artificial intelligence in diagnosing and predicting diabetes (10 studies).
Conclusion: A systematic review of articles revealed that the use of data mining methods for diabetes management in Iran has been associated with good progress, but there is a need to design artificial intelligence systems and algorithms and more measures should be taken in the area of diabetes control and management.
Tomoyuki Ito, Thuy Duong Nguyen, Yutaka Saito
et al.
Despite the advances in surface-display systems for directed evolution, variants with high affinity are not always enriched due to undesirable biases that increase target-unrelated variants during biopanning. Here, our goal was to design a library containing improved variants from the information of the “weakly enriched” library where functional variants were weakly enriched. Deep sequencing for the previous biopanning result, where no functional antibody mimetics were experimentally identified, revealed that weak enrichment was partly due to undesirable biases during phage infection and amplification steps. The clustering analysis of the deep sequencing data from appropriate steps revealed no distinct sequence patterns, but a Bayesian machine learning model trained with the selected deep sequencing data supplied nine clusters with distinct sequence patterns. Phage libraries were designed on the basis of the sequence patterns identified, and four improved variants with target-specific affinity (EC50 = 80–277 nM) were identified by biopanning. The selection and use of deep sequencing data without undesirable bias enabled us to extract the information on prospective variants. In summary, the use of appropriate deep sequencing data and machine learning with the sequence data has the possibility of finding sequence space where functional variants are enriched.
Virtual Mental Health Assistants (VMHAs) are utilized in health care to provide patient services such as counseling and suggestive care. They are not used for patient diagnostic assistance because they cannot adhere to safety constraints and specialized clinical process knowledge (ProKnow) used to obtain clinical diagnoses. In this work, we define ProKnow as an ordered set of information that maps to evidence-based guidelines or categories of conceptual understanding to experts in a domain. We also introduce a new dataset of diagnostic conversations guided by safety constraints and ProKnow that healthcare professionals use (ProKnow-data). We develop a method for natural language question generation (NLG) that collects diagnostic information from the patient interactively (ProKnow-algo). We demonstrate the limitations of using state-of-the-art large-scale language models (LMs) on this dataset. ProKnow-algo incorporates the process knowledge through explicitly modeling safety, knowledge capture, and explainability. As computational metrics for evaluation do not directly translate to clinical settings, we involve expert clinicians in designing evaluation metrics that test four properties: safety, logical coherence, and knowledge capture for explainability while minimizing the standard cross entropy loss to preserve distribution semantics-based similarity to the ground truth. LMs with ProKnow-algo generated 89% safer questions in the depression and anxiety domain (tested property: safety). Further, without ProKnow-algo generations question did not adhere to clinical process knowledge in ProKnow-data (tested property: knowledge capture). In comparison, ProKnow-algo-based generations yield a 96% reduction in our metrics to measure knowledge capture. The explainability of the generated question is assessed by computing similarity with concepts in depression and anxiety knowledge bases. Overall, irrespective of the type of LMs, ProKnow-algo achieved an averaged 82% improvement over simple pre-trained LMs on safety, explainability, and process-guided question generation. For reproducibility, we will make ProKnow-data and the code repository of ProKnow-algo publicly available upon acceptance.
Reconfigurable intelligent surface (RIS) is a revolutionary technology to enhance both the spectral efficiency and energy efficiency of wireless communication systems. However, most of the existing contributions mainly focused on the study of passive RIS, which suffers from the ``double fading'' effect. On the other hand, active RIS, which is equipped with amplifiers, can effectively address this issue. In this paper, we propose an active RIS-aided simultaneous wireless information and power transfer (SWIPT) system. Specifically, we maximize the weighted sum rate of the information receivers, subject to the minimum power received at all energy receivers, amplification power constraint at the active RIS, and the maximum transmit power constraint at the base station (BS). By adopting alternating optimization framework, suboptimal solutions are obtained. Simulation results show that the active RIS-aided SWIPT system has higher performance gain with the same power budget.
In this paper, we consider a multiple-input multiple-output (MIMO) radar system for localizing a target based on its reflected echo signals. Specifically, we aim to estimate the random and unknown angle information of the target, by exploiting its prior distribution information. First, we characterize the estimation performance by deriving the posterior Cramér-Rao bound (PCRB), which quantifies a lower bound of the estimation mean-squared error (MSE). Since the PCRB is in a complicated form, we derive a tight upper bound of it to approximate the estimation performance. Based on this, we analytically show that by exploiting the prior distribution information, the PCRB is always no larger than the Cramér-Rao bound (CRB) averaged over random angle realizations without prior information exploitation. Next, we formulate the transmit signal optimization problem to minimize the PCRB upper bound. We show that the optimal sample covariance matrix has a rank-one structure, and derive the optimal signal solution in closed form. Numerical results show that our proposed design achieves significantly improved PCRB performance compared to various benchmark schemes.
Designing machine learning algorithms that are accurate yet fair, not discriminating based on any sensitive attribute, is of paramount importance for society to accept AI for critical applications. In this article, we propose a novel fair representation learning method termed the Rényi Fair Information Bottleneck Method (RFIB) which incorporates constraints for utility, fairness, and compactness (compression) of representation, and apply it to image and tabular data classification. A key attribute of our approach is that we consider - in contrast to most prior work - both demographic parity and equalized odds as fairness constraints, allowing for a more nuanced satisfaction of both criteria. Leveraging a variational approach, we show that our objectives yield a loss function involving classical Information Bottleneck (IB) measures and establish an upper bound in terms of two Rényi measures of order $α$ on the mutual information IB term measuring compactness between the input and its encoded embedding. We study the influence of the $α$ parameter as well as two other tunable IB parameters on achieving utility/fairness trade-off goals, and show that the $α$ parameter gives an additional degree of freedom that can be used to control the compactness of the representation. Experimenting on three different image datasets (EyePACS, CelebA, and FairFace) and two tabular datasets (Adult and COMPAS), using both binary and categorical sensitive attributes, we show that on various utility, fairness, and compound utility/fairness metrics RFIB outperforms current state-of-the-art approaches.
In recent decades, surveillance and home security systems based on video analysis have been proposed for the automatic detection of abnormal situations. Nevertheless, in several real applications, it may be easier to detect a given event from audio information, and the use of audio surveillance systems can greatly improve the robustness and reliability of event detection. In this paper, a novel system for the detection of polyphonic urban noise is proposed for on-campus audio surveillance. The system aggregates different acoustic features to improve the classification accuracy of urban noise. A combination model composed of a capsule neural network (CapsNet) and recurrent neural network (RNN) is employed as the classifier. CapsNet overcomes some limitations of convolutional neural networks (CNNs), such as the loss of position information after max-pooling, and the RNN mainly models the temporal dependency of context information. The combination of these networks further improves the accuracy and robustness of polyphonic sound events detection. Moreover, a monitoring platform is designed to visualize noise maps and acoustic event information. The deployment architecture of the system is used in real environments, and experiments were also conducted on two public datasets. The results demonstrate that the proposed method is superior to existing state-of-art methods for the polyphonic sound event detection task.
In order to overcome the extremely time-consuming drawback of deep learning (DL), broad learning system (BLS) was proposed as an alternative method. This model is simple, fast, and easy to update. To ensure the fitting and generalization ability of BLS, the hidden layer neurons are often set too many, in fact, a lot of neurons are not needed. Greedy BLS (GBLS) is proposed in this paper to deal with the redundancy of the hidden layer in BLS from another perspective. Different from BLS, the structure of GBLS can be seen as a combination of unsupervised multi-layer feature representation and supervised classification or regression. It trains with a greedy learning scheme, performs principal component analysis (PCA) on the previous hidden layer to form a set of compressed nodes, which are transformed into enhancement nodes and then activated by nonlinear functions. The new hidden layer is composed of all newly generated compressed nodes and enhancement nodes, and so on. The last hidden layer of the network contains the higher-order and abstract essential features of the original data, which is connected to the output layer. Each time a new layer is added to the model, and there is no need to retrain from the beginning, only the previous layer is trained. Experimental results demonstrate that the proposed GBLS model outperforms BLS both in classification and regression.