Incorrect Formation Research Articles

AI-assisted clinical summary and treatment planning for cancer care: A comparative study of human vs. AI-based approaches.

1523 Background: Understanding a patient's clinical narrative, timeline, and history is critical for accurate treatment decision-making. However, reviewing and summarizing complex records is time-consuming and error-prone. Recent advancements in artificial intelligence (AI), specifically large language models (LLM), offer paths to improve quality and efficiency. Methods: A study was conducted on 50 breast cancer cases from an academic medical institution, utilizing all medical records—clinic, pathology, and radiology reports—up until the point of the initial treatment decision. All cases were processed using three different approaches: AI-assisted; full-AI; and human-only. In the AI-assisted method, two oncology physician assistants (PAs) revised AI-generated summaries to create clinical summaries. The full-AI method had AI independently produce clinical summaries, while the human-only method had the PAs compile summaries without AI. Eight board-certified international oncology specialists blindly evaluated summaries for faithfulness, completeness, and succinctness using a 3-point scale, ranked their preferences, and tried to predict which summaries were full-AI. Rankings were assessed using a Friedman test followed by a Wilcoxon signed-rank test, and full-AI prediction was assessed using a two-sided one-sample binomial test. After summarization, a distinct AI system with access to clinical guidelines provided treatment plans. These plans were then evaluated by a board-certified oncologist with access to the original treatment decision. Results: The study found specialists favored AI-assisted, followed by full-AI, and then human-only summaries, with average ranks of 1.73, 1.93, 2.34 respectively (lower is better, p<0.001). The difference between full-AI and AI-assisted was not significant (p=0.11). Evaluation scores (mean±95%CI, higher is better) showed AI-assisted, full-AI, and human-only scored 2.35±0.13, 2.14±0.14, 2.17±0.14 for faithfulness; 2.28±0.12, 2.01±0.12, 1.93±0.14 for completeness; and 2.33±0.12, 2.21±0.12, 1.99±0.13 for succinctness. The average summarization time was 19.71, 1.17, 26.03 minutes. Full-AI identification accuracy was 0.28 (not different from chance 0.33, p=0.46). With AI-assisted summaries, the treatment plans were accurate in 45 cases (90%) and partially accurate in 5 cases (10%). In the 5 partially accurate cases, the system was accurate with the provided input data, but there were inaccuracies with the input data, including incorrect formats or missing data. Conclusions: Incorporating LLMs into the creation of medical summaries has shown improvements in both quality and efficiency, achieving up to 22.2x speed up with full-AI, indicating that AI-assisted summarization tools can potentially enhance care quality. AI-assisted summaries yield accurate treatment plans when the input data is accurate.

The consequences of the use of all kinds of professional activities in the formation of basic educational program for Bachelors in technical areas

The relevance of the research. Two main factors influence the formation of the education process for technical specialties. On one hand, it is the state development strategy and needs of industrial areas. On the other hand, these are the required recommendations of the Ministry of Education for the formation of the educational process of technical specialties. Thus, the Federal educational standard and Professional standard are the two main documents that form the learning process. The problem statement. Incorrect interpretation of these documents can cause an incorrect formation of the main educational program and, as a result, the impossibility of obtaining highly qualified specialists. The purpose of the research is to prove the main errors that can occur when forming the educational process for technical specialties. To solve the problem, the article considers such research methods as literature analysis of, synthesis, generalization. The article examines the transformation of educational standards over the past three decades. The analysis of Federal educational standards and the main educational programs of Bachelors in the direction of training in power supply is carried out. The conclusions. As a result of comparative analysis, possible problems that heads of main educational programs for technical specialties may encounter and their consequences for the educational process have been identified. An approach to the formation of basic educational programs and curriculum has been offered.

Can Large Language Models Understand Real-World Complex Instructions?

Large language models (LLMs) can understand human instructions, showing their potential for pragmatic applications beyond traditional NLP tasks. However, they still struggle with complex instructions, which can be either complex task descriptions that require multiple tasks and constraints, or complex input that contains long context, noise, heterogeneous information and multi-turn format. Due to these features, LLMs often ignore semantic constraints from task descriptions, generate incorrect formats, violate length or sample count constraints, and be unfaithful to the input text. Existing benchmarks are insufficient to assess LLMs’ ability to understand complex instructions, as they are close-ended and simple. To bridge this gap, we propose CELLO, a benchmark for evaluating LLMs' ability to follow complex instructions systematically. We design eight features for complex instructions and construct a comprehensive evaluation dataset from real-world scenarios. We also establish four criteria and develop corresponding metrics, as current ones are inadequate, biased or too strict and coarse-grained. We compare the performance of representative Chinese-oriented and English-oriented models in following complex instructions through extensive experiments. Resources of CELLO are publicly available at https://github.com/Abbey4799/CELLO.

PENERAPAN SAK EMKM DALAM PENYUSUNAN LAPORAN KEUANGAN UMKM PADA CV. MALABO ARSITEK MAKASSAR

This research aims to analyze the implementation of SAK EMKM in the preparation of financial statements for UMKM at CV. Malabo Arsitek Makassar. The data collection method used is triangulation, which is a combination of interviews, observations, and documentation. The types and sources of data used are primary and secondary, while the analysis method follows the approach proposed by Miles and Huberman. Based on the research findings, it can be concluded that the application of SAK EMKM in the preparation of financial statements by CV. Malabo Arsitek does not adhere to the appropriate SAK EMKM standards. This is attributed to the inadequacy of human resources responsible for managing the financial reports of CV. Malabo Arsitek. Additionally, the incorrect format of the financial statements has led to incomplete financial reporting. Therefore, the author recommends that CV. Malabo Arsitek improve the quality of its human resources, especially those involved in financial report management, by studying and understanding SAK EMKM. Furthermore, they should promptly adopt a financial statement format that complies with the applicable standards.

An Excel Template Generator for Darwin Core

Scientific data is diverse and can be complex, potentially including biotic or abiotic measurements, material samples, DNA derived data and more. Especially for researchers who are new to the Darwin Core standard (Darwin Core Task Group 2009), it is not always obvious what the best practice is for creating a Darwin Core Archive (GBIF 2021) for their data. Which core and extensions should they select? Which Darwin Core terms*1 should they include? We present the 'Learnings from Nansen Legacy template generator' (Marsden and Schneider 2023), a spreadsheet template generator to simplify the creation of Darwin Core Archives. It enables users to create a single Microsoft Excel file that includes one sheet per core or extension using a graphical user interface. The user can select from a complete list Darwin Core terms to use as column headers. There are requirements and recommendations for which terms are selected for each core and extension. Descriptions for all terms are displayed when one hovers over a Darwin Core term, and are stored as notes in the template when one select a relevant cell. The generated template includes cell restrictions to prevent one from inputting data in an incorrect format. A separate configuration is also available to aid researchers in creating CF-NetCDF*2 files for physical data, which are also compliant with the FAIR (Findable, Accessible, Interoperable, Resuable)*3 data management principles. The Learnings from Nansen Legacy template generator is published (Marsden and Schneider 2023) and can be installed for use on your website or computer by following the instructions on the software's GitHub repository*4. The template generator can be tested where it is currently hosted, by SIOS*5 (Svalbard Integrated Arctic Earth Observing System). One can also refer to a YouTube tutorial*6 on how the template generator works.

Pelatihan Penyusunan Surat Melalui Mail Merge Bagi Perangkat Gampong Kecamatan Peureulak

This training aims to provide knowledge and skills to gampong officials in Peureulak District in using the mail merge technique to compose letters efficiently and effectively. Participants will learn the steps in creating a template document that includes design patterns, formats, and variable settings that will be used to combine data. Next, participants will learn how to integrate data from external sources such as Excel into Word documents using the mail merge feature. They will learn how to set data filtering criteria, edit and modify documents, and add dynamic elements such as tables, images, and graphs. Training will also cover common troubleshooting techniques related to mail merge, such as resolving data merge errors, fixing incorrect formatting, and optimizing processes to increase productivity. Participants will be involved in hands-on practical sessions where they will be given the opportunity to try out the techniques, they have learned for themselves. This will help them hone their skills and gain confidence in using mail merge in their daily work. This training can provide significant benefits for gampong officials in Peureulak District, enabling them to improve efficiency and accuracy in compiling letters. By using mail merge, the process of writing letters can be automated, saving time and reducing the risk of human error.

A Familial Cohort of Adult-Onset Alexander Disease (P12-3.016)

<h3>Objective:</h3> To describe a familial cohort of adult-onset Alexander Disease. <h3>Background:</h3> Leukodystrophies are a group of rare, hereditary disorders characterized by abnormal myelination. Each leukodystrophy is caused by abnormalities in one or more genes leading to abnormal production of myelin (hypomyelination) or destruction (demyelination). Leukodystrophies are also classified by the region of the brain or in some cases peripheral nerves that are affected. Alexander Disease is a leukodystrophy characterized by a confluent pattern of demyelination. It is caused by autosomal dominant gain-of-function mutation in the <i>GFAP (glial fibrillary acidic protein)</i> gene. <i>GFAP</i> serves an intermediate filament protein of mature astrocytes; incorrect formation of the protein can lead to astrocyte cytoskeletal collapse and abnormal protein aggregation. Clinically, this can lead to innumerable neurologic manifestations in the affected individual. Alexander Disease predominantly affects neonates, infants, children and rarely adults. We describe a familial cohort of adult-onset Alexander Disease associated with a c.758C&gt;G (p.A253G) mutation of GFAP. This mutation has previously been described in neonates. In this case series we describe the clinical and radiologic features of three sisters with adult-onset symptoms. The first patient is a 53-year-old female who presented with progressive myelopathy and later developed bulbar symptoms. Her 51-year-old sister developed progressive, bilateral leg weakness and gait disturbance. She progressed to have profound cognitive dysfunction. The MRI features include T2 hyperintensities with atrophy in the midbrain and upper cervical spine with peripheral rim brainstem and periventricular garlands. <h3>Design/Methods:</h3> N/A <h3>Results:</h3> N/A <h3>Conclusions:</h3> These cases describe the clinical and radiologic features of a familial cohort of adult-onset Alexander disease in a mutation previously described only in childhood onset. This also illustrates the importance of considering a diagnosis of Alexander Disease in adult patients and the necessity of genetic analysis of patients and their family members. <b>Disclosure:</b> Dr. Wolf has nothing to disclose. Dr. Gettings has nothing to disclose. Dr. Goodrich has received personal compensation in the range of $500-$4,999 for serving as a Consultant for EMD Serono.

Calculation of breeding poultry products in the digital economy

The digital economy is penetrating all its industries, including the poultry industry. This creates an objective need to generate more accurate information about the cost of production. The article presents different views of the authors to the concept of «calculation». It has been established that due to the lack of a clear regulation of accounting in many enterprises, the problem of distributing costs between types of manufactured products arises, which leads to incorrect formation of the financial result. It is concluded that it is necessary to develop and implement a methodology for distributing costs between eggs and poultry meat. It is based on the mechanism of transferring losses from the sale of laying hens in a live and slaughtered form to the cost of an egg. The introduction of the proposed methodology will allow to generate more reliable information about the cost.

Frequent linguistic errors in the writing of Yemeni EFL Arabic-speaking learners

The Yemeni EFL learners are prone to share their knowledge and views regarding what and how to say. The constraints of combining this expertise have hampered learners’ writing success. Those obstacles can cause learners to make errors. Error Analysis (EA) and Surface Strategy Taxonomy (SST) were used to analyze learners’ linguistic errors. Error causes were also investigated. This research used a qualitative process style to use a case study approach. Ellis’ five-step EA procedure was followed to analyze essay data each comprising 100-350 words or more written by 20 Yemeni EFL eighth semester Arabic-speaking learners at the Department of Education, Sana’a University, Yemen. They were purposely selected as research subjects. It was noticed that omission was the most common error detected in the learners’ writings. Overall, this form of error accounted for 58.71% of 118 cases out of 201 cases. The learners’ common error categories were the number marker, verb-tenses articles, prepositions, subject-verb agreements, and pronouns. This was preceded by addition (20.39%), incorrect formation (15.92%), and word order (4.97%). Intralingual transfer turned out to be the key reason that caused the errors in the learners’ writing. Any of the interlinguistic comparisons was the cause behind the errors. In terms of verb conjugation component, inflectional morpheme, and auxiliary verb abandonment, Arabic and English have different formal definitions. Interlingual transfer and learning context also caused errors. To prevent errors from fossilizing, language instructors should provide continual corrective feedback, and learners should pursue the correct target language form.

EXPLORING REFERENCE ACCURACY ON UNDERGRADUATE THESIS OF ENGLISH LANGUAGE AND LITERATURE PROGRAM

This study specifically reports on the referencing accuracy to examine the breached APA sixth edition formatting references rules produced by English language and literature undergraduate students. This study was frame in Aronsky et.al (2005) employed a qualitative descriptive method to describe the breaches and the most breached rules. The selection of data samples was obtained from undergraduate thesis of nine students in English language and literature program in a private college in Bandung. The results showed that the students breached six formatting general rules: 1) not on a new page (3.23%), 2) incorrect formation of the section name (29.03%), 3) spacing errors (22.58%), 4) not listed alphabetically (19.35%), 5) indentation errors (16.13%), and 6) in­correct recording of the entries (9.68%). While based on the type of source, 20 rules in total were breached, which spread in six types of sources i.e., 1) periodicals, 2) books, reference books, and book chapters, 3) technical and research reports, 4) doctoral dissertation/ master’s thesis, 5) document from web site, and 6) unclear sources. Furthermore, the most breached APA sixth edition formatting references rules are incorrect formation of the section name (29.03%), incorrect formation of book, reference books, and book chapters (17.44%). The result tracks well with Aronsky et.al (2005) that formatting a paper in APA style can be tricky.

Deep Learning for Multiwell Automatic Log Correction

Researchers have dedicated numerous applications of machine-learning (ML) techniques for fi eld-scale automated interpretation of well-log data. A critical prerequisite for automatic log processing is to ensure that the log characteristics are reasonably consistent across multiple wells. Manually correcting logs for consistency is laborious, subjective, and error prone. For some wellbore logs, such as gamma ray and neutron porosity, borehole effects and miscalibration can cause systematic inconsistencies or errors that might be present even after the application of wellbore and environmental corrections. Biased or consistently inaccurate data in the logs can confound ML approaches into learning erroneous relationships, leading to misinterpretations, such as wrong lithology prediction, reservoir estimation, and incorrect formation markers. To overcome such difficulties, we have developed a deep learning method to provide petrophysicists with a set of consistent logs through the multiwell automatic log correction (MALC) workflow. Presently, the corrections we target are systematic errors on the standard logs, especially gamma ray and neutron logs, random noises, and to a lesser extent, local formation property misreading due to washouts. We applied the proposed method in multiple fi elds worldwide containing different challenges, and in this paper, we include the results in two fi eld examples. The first one covers the correction of synthetic coherent noise added to fi eld data, and the second example covers the correction applied to original measurements.

Investigation of Neural Substrates of Erroneous Behavior in a Delayed-Response Task.

Motor cortical neurons exhibit persistent selective activities (selectivity) during motor planning. Experimental perturbation of selectivity results in the failure of short-term memory retention and consequent behavioral biases, demonstrating selectivity as a neural characteristic of encoding previous sensory input or future action. However, even without experimental manipulation, animals occasionally fail to maintain short-term memory leading to erroneous choice. Here, we investigated neural substrates that lead to the incorrect formation of selectivity during short-term memory. We analyzed neuronal activities in anterior lateral motor cortex (ALM) of mice, a region known to be engaged in motor planning while mice performed the tactile delayed-response task. We found that highly selective neurons lost their selectivity while originally nonselective neurons showed selectivity during the error trials where mice licked toward incorrect direction. We assumed that those alternations would reflect changes in intrinsic properties of population activity. Thus, we estimated an intrinsic manifold shared by neuronal population (shared space), using factor analysis (FA) and measured the association of individual neurons with the shared space by communality, the variance of neuronal activity accounted for by the shared space. We found a positive correlation between selectivity and communality over ALM neurons, which disappeared in erroneous behavior. Notably, neurons showing selectivity alternations between correct and incorrect licking also underwent proportional changes in communality. Our results demonstrated that the extent to which an ALM neuron is associated with the intrinsic manifolds of population activity may elucidate its selectivity and that disruption of this association may alter selectivity, likely leading to erroneous behavior.

Interstitial high dose rate brachytherapy boost in early-stage breast cancer: deformable image registration for defining target volume

Purpose: optimization of the technique of additional irradiation of the removed tumor bed using high-dose brachytherapy for breast cancer. Material and Methods: the results of treatment of 28 patients diagnosed with breast cancer were analyzed. After surgical treatment and a course of external radiation therapy, all patients underwent additional irradiation of the removed tumor bed using high-dose brachytherapy. The assessment of the operation protocols, the data of the pathomorphological conclusion was carried out, and on the basis of pre- and postoperative CT images, the formation of irradiation fields for high-dose brachytherapy was carried out. Results: After deformable (nonrigid) registration of pre- and postoperative CT images of 28 patients, it was revealed that in 18 women (64.3% of cases) the location of interstitial markers and the primary tumor focus does not match topographically, which can cause incorrect formation of borders irradiation. In 35.7% of cases, radiopaque markers were located on the chest wall (on the pectoralis major muscle) when the primary tumor was located in the breast tissue. In 25% of cases, the markers were located cranial or caudal to the topography of the primary tumor focus. Label migration occurred in 3.6% of cases. In 35.7% of cases, the topography of the primary tumor node and marks completely coincided. Conclusions: The use of deformable (non-rigid) registration of pre- and postoperative CT images is a simpler method to determine the topography of the removed tumor bed, which subsequently leads to a more accurate formation of the clinical volume of irradiation.

Independent DSG4 frameshift variants in cats with hair shaft dystrophy

Investigations of hereditary phenotypes in spontaneous mutants may help to better understand the physiological functions of the altered genes. We investigated two unrelated domestic shorthair cats with bulbous swellings of the hair shafts. The clinical, histopathological, and ultrastructural features were similar to those in mice with lanceolate hair phenotype caused by loss-of-function variants in Dsg4 encoding desmoglein 4. We sequenced the genomes from both affected cats and compared the data of each affected cat to 61 control genomes. A search for private homozygous variants in the DSG4 candidate gene revealed independent frameshift variants in each case, c.76del or p.Ile26fsLeu*4 in case no. 1 and c.1777del or p.His593Thrfs*23 in case no. 2. DSG4 is a transmembrane glycoprotein located primarily in the extracellular part of desmosomes, a complex of adhesion molecules responsible for connecting the keratin intermediate filaments of neighbouring epithelial cells. Desmosomes are essential for normal hair shaft formation. Both identified DSG4 variants in the affected cats lead to premature stop codons and truncate major parts of the open-reading frame. We assume that this leads to a complete loss of DSG4 function, resulting in an incorrect formation of the desmosomes and causing the development of defective hair shafts. Together with the knowledge on the effects of DSG4 variants in other species, our data suggest that the identified DSG4 variants cause the hair shaft dystrophy. To the best of our knowledge, this study represents the first report of pathogenic DSG4 variants in domestic animals.

Enhancement of Data Quality in the Information Systems within Organization

Abstract: To introduce unified information systems, this will provide different establishments with an insight on how data related activities take place and there results with assured quality. Considering data accumulation, replication, missing entities, incorrect formatting, anomalies etc. can come to light in the collection of data in different information systems, which can cause an array of adverse effects on data quality, the subject of data quality should be treated with better results. This paper inspects the data quality problems in information systems and introduces the new techniques that enable organizations to improve their quality of data. Keywords: Information Systems (IS), Data Quality, Data Cleaning, Data Profiling, Standardization, Database, Organization

Analysis of Technology Acceptance of Enterprise Resource Planning (ERP) System in The Regional Office of PT. XYZ Throughout Indonesia

The companies that started to renew their system using ERP-based systems, since 2000 the percentage of success reached 75%, and 25% is a failure. There are several issues such as the quality of human resources, not user-friendly system, incorrect format for data recording, system errors, unstable connections, the long-time process in the system, etc. Related to the issue, the ERP SAP technology provider provides local support to help end-users use the system for daily work. In this study, to analyze the factors of acceptance technology of ERP system, we use the UTAUT2 method, and for analyzing those hypotheses, we use the PLS-SEM metho

A Critical Review on the Book “Corona Economy: Resistance Economy in Corona Virus Age”

The unpleasant effects of the Corona illness outbreak on world economies forced some economists to survey and analyze the effects of Corona illness on the economy. The book “Corona Economy: Resistance Economy in Corona Virus Age” is the first significant book in this subject with the purpose of salvation roads recognition for the Iranian economy with resistance economy review. Among the two criticism axes of structure and content, the paper concludes this book as the aggregation of some viewpoints of economic experts, despite many advantages such as innovative and fluent text, it has some defects. The important defects are disconnection in present flow, snap in publishing before determining the Corona effects dimensions, incorrect format, insufficient usage from global studies, and finally lack of comprehensiveness in analysis and solutions. Therefore, it seems the book cannot reach its goals. The paper suggests correction notes to improve and complete its structure and content.

Analysis of Technology Acceptance of Enterprise Resource Planning (ERP) System in The Regional Office of PT. XYZ Throughout Indonesia

The companies that are starting to renew their system using the ERP-based systems, since 2000, the percentage of success reached 75%, and 25% is a failure. The ERP implements have several issues, such as the quality of human resources, not user-friendly system, incorrect format for data recording, system errors, unstable connections, a long-time process in the system, etc. Related to the issue, the company, which is a provider in the ERP SAP marketing, assists by local support to help end-users regarding the recording of selected divisions that use this system for daily work. It makes the researcher do the analysis related to what factors influence the ERP SAP acceptance technology with local support or implementor as a support for users of those companies in the regional office. In this study, to analyze the ERP system’s acceptance technology factors, we use the UTAUT2 method with two additional constructed variables, Trust and Learning Value, and for analyzing those hypotheses, we use the PLS-SEM method. The study results show that several variable factors significantly influence the technology acceptance of the ERP SAP in PT. XYZ, which is Hedonic Motivation, Price Value, and Habit. Local support associated with Effort Expectancy and Social Influence variables does not significantly influence the system’s acceptance. Researchers recommend that three factors should be considered more closely by the ERP system providers as a material reference that is related to the further strategy development and system development, such as improving the quality of human resources, etc.

MODELING THE SYNTHESIS OF TAKAGI — SUGENO — KANG FUZZY CONTROLLERS IN SOME CONTROL SYSTEMS

Modern challenges in a post-industrial society require further development of management systems for complex technical and technological phenomena and processes. Effective control of an object is possible if a controller, or a fuzzy controller, correctly generates the required control action. Recently, fuzzy controllers have been very popular. Fuzzy logical statements in this case help considering various nonlinear relationships. The synthesis of the fuzzy controller parameters allows for more efficient operation of the control system. A possible option for obtaining the best set of parameters for a fuzzy controller is the use of genetic algorithms for its synthesis. The use of genetic algorithms for the fuzzy controllers synthesis can lead to the fact that the elements of its parameters array will change in such a way that an incorrect value of one or more elements will occur. This situation leads to impossibility of composing membership functions for the terms of the variables of the fuzzy controller. Incorrect value formation is excluded by constructing a limited functional dependency. This paper proposes a mathematical model of the parameters of the term-set of variables of a fuzzy controller of the Takagi — Sugeno — Kang type of the zero and first orders. The authors disclose the content of the conditions and conclusions of the rule base for the fuzzy controller of the above type. As a result of the simulation, some operations of the genetic algorithm are implemented using a random number generator. Graphical models of the membership functions of the input variables of the fuzzy controller of the type under consideration clearly illustrate the occurrence of all parameters in their range of possible values. A description of the control system operation with two control parameters and one control action at the specified values of the control parameters is presented.

УДОСКОНАЛЕННЯ АКАДЕМІЧНОГО МОВЛЕННЯ ЗДОБУВАЧІВ ОСВІТНЬОГО СТУПЕНЯ «ДОКТОР ФІЛОСОФІЇ»

The article is devoted to the elucidation of ways to improve the academic speech of applicants for the degree of "Doctor of Philosophy". The concept of "academic speech" is understood as the activity of the researcher to create specialized professional scientific texts and present orally or in writing the results of research at scientific events of various levels. It is noted that the improvement of academic speech in the conditions of postgraduate / adjunct studies can be significantly contributed by the discipline "Fundamentals of Academic Speech", which aims to deepen knowledge of modern Ukrainian literary language, to develop a sense of scientific language, the language taste, to master the scientific style of speech and writing, language means of the scientific text, to develop the steady abilities and skills of the correct and communicatively justified use of language means in academic speech, to form the speech behavior in oral scientific discussion. The most common mistakes that occur in scientific texts are identified. They include: lexical (tracings from the Russian language, tautology, pleonasms, incorrect use of paronyms), grammatical (incorrect formation of singular forms of nouns, adjectives, numerals, pronouns, incorrect use of prepositions, incorrect formation of participles, incorrect use of pronouns), syntactic (oversaturation of the scientific text with the verb forms on "-ся", non-normative use of prepositions as a part of syntactic constructions); stylistic, punctuation, spelling mistakes. A number of exercises, methods and forms of organization of the educational process which can contribute to the improvement of academic speech of future scientists are proposed. They include exercises on editing and translating scientific texts, role-playing and business games, discussions, debates, brainstorming, project method, preparation of own scientific texts, speeches with an oral scientific report in compliance with the requirements for public speaking.