Search

Recent years have witnessed the resurgence of knowledge engineering which is featured by the fast growth of knowledge graphs. However, most of existing knowledge graphs are represented with pure symbols, which hurts the machine's capability to understand the real world. The multi-modalization of knowledge graphs is an inevitable key step towards the realization of human-level machine intelligence. The results of this endeavor are Multi-modal Knowledge Graphs (MMKGs). In this survey on MMKGs constructed by texts and images, we first give definitions of MMKGs, followed with the preliminaries on multi-modal tasks and techniques. We then systematically review the challenges, progresses and opportunities on the construction and application of MMKGs respectively, with detailed analyses of the strengths and weaknesses of different solutions. We finalize this survey with open research problems relevant to MMKGs.

Document-level Relation Extraction (RE) is a promising task aiming at identifying relations of multiple entity pairs in a document. Compared with the sentence-level counterpart, it has raised two significant challenges: a) In most cases, a relational fact can be adequately expressed via a small subset of sentences from the document, namely evidence. But the traditional method cannot model such strong semantic correlations between evidence sentences that collaborate to describe a specific relation; b) The data of this task is extremely long-tail in terms of too many NA instances and imbalanced relational types. Such data can mislead the tail prediction bias to the head categories in the RE model. In this paper, we present a novel Evidence reasoning and Curriculum learning method for DocRE (DRE-EC) to address these challenges. Particularly, we first formulate evidence extraction as a sequential decision problem through a crafted reinforcement learning mechanism with an efficient path searching strategy to reduce the action space. Providing the evidence for each entity pair as a customized-filtered document in advance helps infer the relations better. To address the long-tail issue, we further develop a hybrid curriculum learning method at the NA-level (NC) and relation-level (RC) with our customized difficulty measure score. In NC, the NA samples are scheduled in an easy-to-hard scheme and gradually added, resulting in the data distribution from ideal and balanced to real and unbalanced. In RC, the scheme is switched into hard-to-easy to enhance the hard and tail samples. In addition, we propose a new Equalization adaptive Focal Loss(EFLoss) that can adjust to the changing data distribution and focus more on the tail categories. We conduct various experiments on two document-level RE benchmarks and achieve a remarkable improvement over previous competitive baselines. Furthermore, we provide detailed analyses of the advantages and effectiveness of our method.

Crystallization, as the final step of production, has a significant impact on the crystal properties, including particle size distribution (PSD), crystal habit, and crystal form. By applying the process analysis technique (PAT), the crystallization process can be monitored and the crystallization parameters can be regulated to obtain higher quality crystals. This work demonstrates the development and validation of a Raman spectroscopy concentration calibration model for monitoring the dl-methionine solute concentration. The results show that the Raman spectroscopy concentration calibration model can obtain an accurate concentration change of the crystallization process, and the relative standard deviations of the repeatability and intermediate precision were less than 5%. Then, concentration variations of the crystallization process obtained by Raman spectroscopy were used to determine the exact crystallization kinetic parameters of β-dl-methionine by solving the population balance equation (PBE). After that, the PSD under other conditions was simulated by inputting kinetic data into the PBE. Moreover, the impact of the crystallization operation parameters on the product particle size was investigated. It was observed that either increasing the temperature or supersaturation of the crystallization process could increase the particle size of the crystallized product.

Question understanding is an important issue to the success of a Knowledge-based Question Answering (KBQA) system.However, the existing study does not pay enough attention to this issue given that the questions in the existing KBQA datasets are usually expressed in simple and straightforward way. This is not in line with the actual linguistic conventions, which often use a lot of modifiers. To facilitate the study on evaluating and enhancing the question understanding ability of the KBQA systems, this paper proposes to construct a complex-modified question-answering (XMQAs) dataset based on existing KBQA datasets. With the help of knowledge bases and dictionaries, three kinds of modifiers are defined and applied to original simple-expressed questions. These modifiers could make the expression of these questions complex without changing their semantics. Based on XMQAs, we then propose a novel question understanding algorithm upon existing KBQA models, which greatly improves the robustness of their question understanding abilities. We conduct extensive experiments on XMQAs and two widely acknowledged KBQA datasets. The empirical results demonstrate that our proposed algorithm can improve the performance of KBQA models on not only the complex-modified questions, but also simple-expressed questions.

The solubility of 4-chlorophenoxyacetic acid (4-CPA) in 13 monosolvents including ethanol, 1-propanol, isopropanol, 1-butanol, 2-butanol, isobutanol, ethyl formate, methyl acetate, ethyl acetate, propyl acetate, butyl acetate, acetonitrile, and 1,4-dioxane was measured by a gravimetric method from 283.15 to 328.15 K at an atmospheric pressure of 0.1013 MPa. The solubility of 4-CPA increased with temperature and followed the order 1,4-dioxane > alcohols > esters > acetonitrile at all experimental temperatures. The solvent effects, especially solvent polarity, hydrogen bond donor propensity, and hydrogen bond acceptor propensity, were used to analyze the solubility of 4-CPA in different solvents. The solubility basically follows the rule “like dissolves like” and has a higher value in solvents with strong hydrogen bond donor propensity and hydrogen bond acceptor propensity. Besides, the modified Apelblat equation, λhequation, and van’t Hoff equation were applied to correlate the measured solubility data. The average relative deviation (ARD) values showed that the correlation results of the three models agree well with the experimental data.

With the rapid development of location-based services, a particularly important aspect of start-up marketing research is to explore and characterize points of interest (PoIs) such as restaurants and hotels on maps. However, due to the lack of direct access to PoI databases, it is necessary to rely on existing APIs to query PoIs within a region and calculate PoI statistics. Unfortunately, public APIs generally impose a limit on the maximum number of queries. Therefore, we propose effective and efficient sampling methods based on road networks to sample PoIs on maps and provide unbiased estimators for calculating PoI statistics. In general, the more intense the roads, the denser the distribution of PoIs is within a region. Experimental results show that compared with state-of-the-art methods, our sampling methods improve the efficiency of aggregate statistical estimations.

Some polycrystalline drugs are subject to concomitant polymorphism during the manufacturing process, resulting in large batch-to-batch variation and reduced drug bioavailability. Most existing studies on concomitant polymorphism have focused only on differences in nucleation kinetics and lacked analysis of the role of growth processes in polymorph control. Herein, we conducted a systematic study on the concomitant crystallization of aripiprazole Form III and Form V under the influence of thermodynamic and kinetic factors employing thermal analysis with the Ostwald ratio as a medium. It shows that the nucleation and growth rates of both forms cross over with increasing supersaturation. The comparison between the polymorph composition computed by the Ostwald ratio and the experiments highlights the considerable influence of crystal growth on the formation of concomitant polymorphism. The effect of crystal growth at lower supersaturation is indeed minimal due to the large difference in nucleation rates, but as the supersaturation increases, the difference in nucleation rates becomes smaller and the growth rate ratio gradually becomes a key factor. The images plotted with initial concentration and crystallization temperature further demonstrate that the introduction of actual nonlinear growth kinetics can lead to improved computation accuracy of the Ostwald ratio. This work provides evidence for the application of the Ostwald ratio in the design of batch crystallization for a concomitant system.

A gravimetric method is adopted to measure the solid–liquid equilibrium of 3-amino-2-methylbenzoic acid in different solvents, including methanol, ethanol, 1-propanol, isopropyl alcohol, 1-butanol, isobutyl alcohol, 2-butanol, methyl acetate, ethyl acetate, propyl acetate, water, and acetonitrile. Moreover, the experiment is carried out at atmospheric pressure, and the temperature range is from 288.15 to 328.15 K. The solid–liquid equilibrium of 3-amino-2-methylbenzoic acid in all used solvents increases with increasing temperature. In alcohol solvents, except for isopropyl alcohol and 2-butanol, the solubility of 3-amino-2-methylbenzoic acid is positively related to the polarity of the solvent, which follows the “like dissolves like” rule. Because isopropyl alcohol and 2-butanol possess a high hydrogen bond acceptor propensity, this leads to a different solubility behavior. In nonalcohol solvents, the solid–liquid equilibrium of 3-amino-2-methylbenzoic acid also follows the “like dissolves like” rule. Among all 12 solvents, the solid–liquid equilibria of 3-amino-2-methylbenzoic acid in water and acetonitrile are lower than those in alcohols and esters. In addition, the solid–liquid equilibrium data of 3-amino-2-methylbenzoic acid were related by the modified Apelblat equation, van’t Hoff equation, and λhequation. Furthermore, all relevant results show consistency with the solid–liquid equilibrium data of 3-amino-2-methylbenzoic acid.