Search

The impact of research papers, typically measured in terms of citation counts, depends on several factors, including the reputation of the authors, journals, and institutions, in addition to the quality of the scientific work. In this paper, we present an approach that combines natural language processing and machine learning to predict the impact of papers in a specific journal. Our focus is on the text, which should correlate with impact and the topics covered in the research. We employed a dataset of over 40,000 articles from ACS Applied Materials and Interfaces spanning from 2012 to 2022. The data was processed using various text embedding techniques and classified with supervised machine learning algorithms. Papers were categorized into the top 20% most cited within the journal, using both yearly and cumulative citation counts as metrics. Our analysis reveals that the method employing generative pre-trained transformers (GPT) was the most efficient for embedding, while the random forest algorithm exhibited the best predictive power among the machine learning algorithms. An optimized accuracy of 80\% in predicting whether a paper was among the top 20% most cited was achieved for the cumulative citation count when abstracts were processed. This accuracy is noteworthy, considering that author, institution, and early citation pattern information were not taken into account. The accuracy increased only slightly when the full texts of the papers were processed. Also significant is the finding that a simpler embedding technique, term frequency-inverse document frequency (TFIDF), yielded performance close to that of GPT. Since TFIDF captures the topics of the paper we infer that, apart from considering author and institution biases, citation counts for the considered journal may be predicted by identifying topics and "reading" the abstract of a paper.

Herein we review aspects of leading-edge research and innovation in chemistry which exploits big data and machine learning (ML), two computer science fields that combine to yield machine intelligence. ML can accelerate the solution of intricate chemical problems and even solve problems that otherwise would not be tractable. But the potential benefits of ML come at the cost of big data production; that is, the algorithms, in order to learn, demand large volumes of data of various natures and from different sources, from materials properties to sensor data. In the survey, we propose a roadmap for future developments, with emphasis on materials discovery and chemical sensing, and within the context of the Internet of Things (IoT), both prominent research fields for ML in the context of big data. In addition to providing an overview of recent advances, we elaborate upon the conceptual and practical limitations of big data and ML applied to chemistry, outlining processes, discussing pitfalls, and reviewing cases of success and failure.

Well-established automatic analyses of texts mainly consider frequencies of linguistic units, e.g. letters, words and bigrams, while methods based on co-occurrence networks consider the structure of texts regardless of the nodes label (i.e. the words semantics). In this paper, we reconcile these distinct viewpoints by introducing a generalized similarity measure to compare texts which accounts for both the network structure of texts and the role of individual words in the networks. We use the similarity measure for authorship attribution of three collections of books, each composed of 8 authors and 10 books per author. High accuracy rates were obtained with typical values from 90% to 98.75%, much higher than with the traditional the TF-IDF approach for the same collections. These accuracies are also higher than taking only the topology of networks into account. We conclude that the different properties of specific words on the macroscopic scale structure of a whole text are as relevant as their frequency of appearance; conversely, considering the identity of nodes brings further knowledge about a piece of text represented as a network.

Mild Cognitive Impairment (MCI) is a mental disorder difficult to diagnose. Linguistic features, mainly from parsers, have been used to detect MCI, but this is not suitable for large-scale assessments. MCI disfluencies produce non-grammatical speech that requires manual or high precision automatic correction of transcripts. In this paper, we modeled transcripts into complex networks and enriched them with word embedding (CNE) to better represent short texts produced in neuropsychological assessments. The network measurements were applied with well-known classifiers to automatically identify MCI in transcripts, in a binary classification task. A comparison was made with the performance of traditional approaches using Bag of Words (BoW) and linguistic features for three datasets: DementiaBank in English, and Cinderella and Arizona-Battery in Portuguese. Overall, CNE provided higher accuracy than using only complex networks, while Support Vector Machine was superior to other classifiers. CNE provided the highest accuracies for DementiaBank and Cinderella, but BoW was more efficient for the Arizona-Battery dataset probably owing to its short narratives. The approach using linguistic features yielded higher accuracy if the transcriptions of the Cinderella dataset were manually revised. Taken together, the results indicate that complex networks enriched with embedding is promising for detecting MCI in large-scale assessments, Comment: Published in Annual Meeting of the Association for Computational Linguist 2017

The identification of authorship in disputed documents still requires human expertise, which is now unfeasible for many tasks owing to the large volumes of text and authors in practical applications. In this study, we introduce a methodology based on the dynamics of word co-occurrence networks representing written texts to classify a corpus of 80 texts by 8 authors. The texts were divided into sections with equal number of linguistic tokens, from which time series were created for 12 topological metrics. The series were proven to be stationary (p-value>0.05), which permits to use distribution moments as learning attributes. With an optimized supervised learning procedure using a Radial Basis Function Network, 68 out of 80 texts were correctly classified, i.e. a remarkable 85% author matching success rate. Therefore, fluctuations in purely dynamic network metrics were found to characterize authorship, thus opening the way for the description of texts in terms of small evolving networks. Moreover, the approach introduced allows for comparison of texts with diverse characteristics in a simple, fast fashion.

The use of science to understand its own structure is becoming popular, but understanding the organization of knowledge areas is still limited because some patterns are only discoverable with proper computational treatment of large-scale datasets. In this paper, we introduce a network-based methodology combined with text analytics to construct the taxonomy of science fields. The methodology is illustrated with application to two topics: complex networks (CN) and photonic crystals (PC). We built citation networks using data from the Web of Science and used a community detection algorithm for partitioning to obtain science maps of the fields considered. We also created an importance index for text analytics in order to obtain keywords that define the communities. A dendrogram of the relatedness among the subtopics was also obtained. Among the interesting patterns that emerged from the analysis, we highlight the identification of two well-defined communities in PC area, which is consistent with the known existence of two distinct communities of researchers in the area: telecommunication engineers and physicists. With the methodology, it was also possible to assess the interdisciplinary and time evolution of subtopics defined by the keywords. The automatic tools described here are potentially useful not only to provide an overview of scientific areas but also to assist scientists in performing systematic research on a specific topic.

The representation of basic elements of music in terms of discrete audio signals is often used in software for musical creation and design. Nevertheless, there is no unified approach that relates these elements to the discrete samples of digitized sound. In this article, each musical element is related by equations and algorithms to the discrete-time samples of sounds, and each of these relations are implemented in scripts within a software toolbox, referred to as MASS (Music and Audio in Sample Sequences). The fundamental element, the musical note with duration, volume, pitch and timbre, is related quantitatively to characteristics of the digital signal. Internal variations of a note, such as tremolos, vibratos and spectral fluctuations, are also considered, which enables the synthesis of notes inspired by real instruments and new sonorities. With this representation of notes, resources are provided for the generation of higher scale musical structures, such as rhythmic meter, pitch intervals and cycles. This framework enables precise and trustful scientific experiments, data sonification and is useful for education and art. The efficacy of MASS is confirmed by the synthesis of small musical pieces using basic notes, elaborated notes and notes in music, which reflects the organization of the toolbox and thus of this article. It is possible to synthesize whole albums through collage of the scripts and settings specified by the user. With the open source paradigm, the toolbox can be promptly scrutinized, expanded in co-authorship processes and used with freedom by musicians, engineers and other interested parties. In fact, MASS has already been employed for diverse purposes which include music production, artistic presentations, psychoacoustic experiments and computer language diffusion where the appeal of audiovisual artifacts is exploited for education., Comment: A software toolbox, a Python Package, musical pieces and further documents are in: https://github.com/ttm/mass

The development of new statistical and computational methods is increasingly making it possible to bridge the gap between hard sciences and humanities. In this study, we propose an approach based on a quantitative evaluation of attributes of objects in fields of humanities, from which concepts such as dialectics and opposition are formally defined mathematically. As case studies, we analyzed the temporal evolution of classical music and philosophy by obtaining data for 8 features characterizing the corresponding fields for 7 well-known composers and philosophers, which were treated with multivariate statistics and pattern recognition methods. A bootstrap method was applied to avoid statistical bias caused by the small sample data set, with which hundreds of artificial composers and philosophers were generated, influenced by the 7 names originally chosen. Upon defining indices for opposition, skewness and counter-dialectics, we confirmed the intuitive analysis of historians in that classical music evolved according to a master-apprentice tradition, while in philosophy changes were driven by opposition. Though these case studies were meant only to show the possibility of treating phenomena in humanities quantitatively, including a quantitative measure of concepts such as dialectics and opposition the results are encouraging for further application of the approach presented here to many other areas, since it is entirely generic., Comment: arXiv admin note: substantial text overlap with arXiv:1109.4653

Concepts and methods of complex networks have been employed to uncover patterns in a myriad of complex systems. Unfortunately, the relevance and significance of these patterns strongly depends on the reliability of the data sets. In the study of collaboration networks, for instance, unavoidable noise pervading author's collaboration datasets arises when authors share the same name. To address this problem, we derive a hybrid approach based on authors' collaboration patterns and on topological features of collaborative networks. Our results show that the combination of strategies, in most cases, performs better than the traditional approach which disregards topological features. We also show that the main factor for improving the discriminability of homonymous authors is the average distance between authors. Finally, we show that it is possible to predict the weighting associated to each strategy compounding the hybrid system by examining the discrimination obtained from the traditional analysis of collaboration patterns. Once the methodology devised here is generic, our approach is potentially useful to classify many other networked systems governed by complex interactions., Comment: To appear in Scientometrics, 2014

This paper reports on stable (or invariant) properties of human interaction networks, with benchmarks derived from public email lists. Activity, recognized through messages sent, along time and topology were observed in snapshots in a timeline, and at different scales. Our analysis shows that activity is practically the same for all networks across timescales ranging from seconds to months. The principal components of the participants in the topological metrics space remain practically unchanged as different sets of messages are considered. The activity of participants follows the expected scale-free trace, thus yielding the hub, intermediary and peripheral classes of vertices by comparison against the Erd\"os-R\'enyi model. The relative sizes of these three sectors are essentially the same for all email lists and the same along time. Typically, $<15\%$ of the vertices are hubs, 15-45\% are intermediary and $>45\%$ are peripheral vertices. Similar results for the distribution of participants in the three sectors and for the relative importance of the topological metrics were obtained for 12 additional networks from Facebook, Twitter and ParticipaBR. These properties are consistent with the literature and may be general for human interaction networks, which has important implications for establishing a typology of participants based on quantitative criteria., Comment: See ancillary Supporting Information PDF file for further tables and figures. More information on code and further files can be found at https://github.com/ttm/articleStabilityInteractionNetworks

The identification of modular structures is essential for characterizing real networks formed by a mesoscopic level of organization where clusters contain nodes with a high internal degree of connectivity. Many methods have been developed to unveil community structures, but only a few studies have probed their suitability in incomplete networks. Here we assess the accuracy of community detection techniques in incomplete networks generated in sampling processes. We show that the walktrap and fast greedy algorithms are highly accurate for detecting the modular structure of incomplete complex networks even if many of their nodes are removed. Furthermore, we implemented an approach that improved the time performance of the walktrap and fast greedy algorithms, while retaining the accuracy rate in identifying the community membership of nodes. Taken together our results show that this new approach can be applied to speed up virtually any community detection method in dense complex networks, as it is the case of similarity networks.

While the use of statistical physics methods to analyze large corpora has been useful to unveil many patterns in texts, no comprehensive investigation has been performed investigating the properties of statistical measurements across different languages and texts. In this study we propose a framework that aims at determining if a text is compatible with a natural language and which languages are closest to it, without any knowledge of the meaning of the words. The approach is based on three types of statistical measurements, i.e. obtained from first-order statistics of word properties in a text, from the topology of complex networks representing text, and from intermittency concepts where text is treated as a time series. Comparative experiments were performed with the New Testament in 15 different languages and with distinct books in English and Portuguese in order to quantify the dependency of the different measurements on the language and on the story being told in the book. The metrics found to be informative in distinguishing real texts from their shuffled versions include assortativity, degree and selectivity of words. As an illustration, we analyze an undeciphered medieval manuscript known as the Voynich Manuscript. We show that it is mostly compatible with natural languages and incompatible with random texts. We also obtain candidates for key-words of the Voynich Manuscript which could be helpful in the effort of deciphering it. Because we were able to identify statistical measurements that are more dependent on the syntax than on the semantics, the framework may also serve for text analysis in language-dependent applications.

There are different ways to define similarity for grouping similar texts into clusters, as the concept of similarity may depend on the purpose of the task. For instance, in topic extraction similar texts mean those within the same semantic field, whereas in author recognition stylistic features should be considered. In this study, we introduce ways to classify texts employing concepts of complex networks, which may be able to capture syntactic, semantic and even pragmatic features. The interplay between the various metrics of the complex networks is analyzed with three applications, namely identification of machine translation (MT) systems, evaluation of quality of machine translated texts and authorship recognition. We shall show that topological features of the networks representing texts can enhance the ability to identify MT systems in particular cases. For evaluating the quality of MT texts, on the other hand, high correlation was obtained with methods capable of capturing the semantics. This was expected because the golden standards used are themselves based on word co-occurrence. Notwithstanding, the Katz similarity, which involves semantic and structure in the comparison of texts, achieved the highest correlation with the NIST measurement, indicating that in some cases the combination of both approaches can improve the ability to quantify quality in MT. In authorship recognition, again the topological features were relevant in some contexts, though for the books and authors analyzed good results were obtained with semantic features as well. Because hybrid approaches encompassing semantic and topological features have not been extensively used, we believe that the methodology proposed here may be useful to enhance text classification considerably, as it combines well-established strategies.

Many features of complex systems can now be unveiled by applying statistical physics methods to treat them as social networks. The power of the analysis may be limited, however, by the presence of ambiguity in names, e.g., caused by homonymy in collaborative networks. In this paper we show that the ability to distinguish between homonymous authors is enhanced when longer-distance connections are considered, rather than looking at only the immediate neighbors of a node in the collaborative network. Optimized results were obtained upon using the 3rd hierarchy in connections. Furthermore, reasonable distinction among authors could also be achieved upon using pattern recognition strategies for the data generated from the topology of the collaborative network. These results were obtained with a network from papers in the arXiv repository, into which homonymy was deliberately introduced to test the methods with a controlled, reliable dataset. In all cases, several methods of supervised and unsupervised machine learning were used, leading to the same overall results. The suitability of using deeper hierarchies and network topology was confirmed with a real database of movie actors, with the additional finding that the distinguishing ability can be further enhanced by combining topology features and long-range connections in the collaborative network.

Methods from statistical physics, such as those involving complex networks, have been increasingly used in quantitative analysis of linguistic phenomena. In this paper, we represented pieces of text with different levels of simplification in co-occurrence networks and found that topological regularity correlated negatively with textual complexity. Furthermore, in less complex texts the distance between concepts, represented as nodes, tended to decrease. The complex networks metrics were treated with multivariate pattern recognition techniques, which allowed us to distinguish between original texts and their simplified versions. For each original text, two simplified versions were generated manually with increasing number of simplification operations. As expected, distinction was easier for the strongly simplified versions, where the most relevant metrics were node strength, shortest paths and diversity. Also, the discrimination of complex texts was improved with higher hierarchical network metrics, thus pointing to the usefulness of considering wider contexts around the concepts. Though the accuracy rate in the distinction was not as high as in methods using deep linguistic knowledge, the complex network approach is still useful for a rapid screening of texts whenever assessing complexity is essential to guarantee accessibility to readers with limited reading ability, Comment: The Supplementary Information (SI) is available from https://dl.dropbox.com/u/2740286/supplementary.pdf

The automatic disambiguation of word senses (i.e., the identification of which of the meanings is used in a given context for a word that has multiple meanings) is essential for such applications as machine translation and information retrieval, and represents a key step for developing the so-called Semantic Web. Humans disambiguate words in a straightforward fashion, but this does not apply to computers. In this paper we address the problem of Word Sense Disambiguation (WSD) by treating texts as complex networks, and show that word senses can be distinguished upon characterizing the local structure around ambiguous words. Our goal was not to obtain the best possible disambiguation system, but we nevertheless found that in half of the cases our approach outperforms traditional shallow methods. We show that the hierarchical connectivity and clustering of words are usually the most relevant features for WSD. The results reported here shine light on the relationship between semantic and structural parameters of complex networks. They also indicate that when combined with traditional techniques the complex network approach may be useful to enhance the discrimination of senses in large texts, Comment: The Supplementary Information (SI) is available from http://dl.dropbox.com/u/2740286/epl_SI.pdf

The use of statistical methods to analyze large databases of text has been useful to unveil patterns of human behavior and establish historical links between cultures and languages. In this study, we identify literary movements by treating books published from 1590 to 1922 as complex networks, whose metrics were analyzed with multivariate techniques to generate six clusters of books. The latter correspond to time periods coinciding with relevant literary movements over the last 5 centuries. The most important factor contributing to the distinction between different literary styles was {the average shortest path length (particularly, the asymmetry of the distribution)}. Furthermore, over time there has been a trend toward larger average shortest path lengths, which is correlated with increased syntactic complexity, and a more uniform use of the words reflected in a smaller power-law coefficient for the distribution of word frequency. Changes in literary style were also found to be driven by opposition to earlier writing styles, as revealed by the analysis performed with geometrical concepts. The approaches adopted here are generic and may be extended to analyze a number of features of languages and cultures., Comment: The Supplementary Information (SI) is available from http://iopscience.iop.org/1367-2630/14/4/043029/media/njp043029suppdata.pdf

In this paper we quantify the consistency of word usage in written texts represented by complex networks, where words were taken as nodes, by measuring the degree of preservation of the node neighborhood.} Words were considered highly consistent if the authors used them with the same neighborhood. When ranked according to the consistency of use, the words obeyed a log-normal distribution, in contrast to the Zipf's law that applies to the frequency of use. Consistency correlated positively with the familiarity and frequency of use, and negatively with ambiguity and age of acquisition. An inspection of some highly consistent words confirmed that they are used in very limited semantic contexts. A comparison of consistency indices for 8 authors indicated that these indices may be employed for author recognition. Indeed, as expected authors of novels could be distinguished from those who wrote scientific texts. Our analysis demonstrated the suitability of the consistency indices, which can now be applied in other tasks, such as emotion recognition.