Skip to content
Free planNo credit card requiredStart writing
AI for Research home
AI for Research home
Dashboard
Sign inGet started

Search 250 million research papers

Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.

About 10,000 papers for “Machine code”

Multi-lingual code-mixed machine translation system

Amit Ramesh Khaparde, Akshat Kumar, Mahesh Kumar et al. · AIP conference proceedings · 2022

The Machine Translation System has been lately attracting considerable recognition due to its extensive use in multiple fields. This study focuses on abundant raw code-mixed textual data on microblogging sites, which can't be processed directly because it's multilingual. This raises a language barrier in communication and research. In order to use this immense data source, all unwanted symbols are removed to reduce noise in the input text. Also, the emoticon adding up the intention or emotion is permuted with the corresponding textual meaning. Recurrent Neural Network in one form or the other can be used for translating multilingual text from source to target language. A large neural network is used in the process of generating the expected noise-free output in the targeted language. The extensive experiments demonstrate that removing noise and transmuted emoticons improve the result and sustain the intention or semantics better. The proposed system can save anywhere from 35% to 98% of the cost of a human translator because of higher accuracy and desired output. It also finds a place as a subsystem to real-time multi-lingual translators, manuscript translators, other systems for providing input as required targeted language thus enlarging to input domain for that system.

Cited by 1

Converting Python Virtual Machine Code to C

John Aycock · 1998

The optimization of a Python program has a limit point, beyond which a programmer must resort to C code in order to get more speed. Not all programmers are willing or able to take this step. 211 is an experimental program which automatically converts Python virtual machine code into C. In this paper I discuss 211, its results, and suggest changes to Python's internals which should yield better results and faster programs. 1 Introduction My first nontrivial Python program was a 1400-line program that built and manipulated large graphs. While I found Python made for very fast development, the result was somewhat disappointing --- medium-sized inputs could take days to run, making large inputs out of the question. I applied the usual optimization techniques [3]. Using Python's profiler, I was able to identify "hot spots" in the program where the majority of time was being spent; changing that code to use more efficient data structures and cache previouslycomputed information sped the pro...

Cited by 0

Chapter 13: Sub-machine-code Genetic Programming

Riccardo Poli, William B Langdon, William B · 1999

Introduction Genetic Programming (GP) [Koza, 1992; Koza, 1994; Banzhaf et al., 1998] is usually seen as quite demanding from the computation load and memory use point of view. So, over the years a number of ideas on how to improve GP performance have been proposed in the literature. We recall the main speedup techniques published to date in Section 13.2. Some of these techniques are now used in many GP implementations. Thanks to this and to the fact that the power of our workstations is increasing exponentially (today's CPUs are now more than 10 times faster than those used in early GP work), nowadays we can run 50 generations a typical GP benchmark problem with a population of 500 individuals in perhaps ten seconds on a normal workstation. Nonetheless, the demand for more and more efficient implementations has not stopped. This is because extensive experimental GP studies (like [Langdon and Poli, 1998] or [Luke and Spector, 1998]) and complex applications (like [Poli, 1996]

Cited by 0

Generation of reconfigurable circuits from machine code

Nuno Paulino · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · 2011

Tese de mestrado integrado. Engenharia Electrotécnica e de Computadores. Telecomunicações. Universidade do Porto. Faculdade de Engenharia. 2011

Cited by 0Open access

Study of Machine Learning No-Code Platform

Kartikey Ahlawat (13880598) · INDIGO (University of Illinois at Chicago) · 2022

<p><em><strong>Study on a web application based on no-code platform ML Modeller, which targets to provide predictions using various machine learning algorithms under classification and regression. Users have to upload data on which predictions are needed, the no-code platform will automatically train on the data and even show you the accuracy is achieved, now users can get required predictions. No-Code platforms will reduce Manual Data Analyzing efforts and overcome the problem of false insights and guessing. Website’s performance is also evaluated if people are using it or not</strong></em> </p>

Cited by 0Open access

Machine Learning aided Linux Kernel Code Analysis

Krogerus, Heikki · Theseus (Ammattikorkeakoulujen) · 2024

Jokainen koodimuutos, jota ehdotetaan Linux-käyttöjärjestelmäytimeen, testataan monilla eri
\nautomaattisilla testeillä. Tästä huolimatta jokainen muutos on vaatinut ihmisten tekemän
\nkatselmoinnin. Johtuen rajallisesta määrästä ihmisiä, jotka pystyvät katselmoimaan Linux koodia,
\nmonia virheitä on jäänyt huomaamatta.
\nAutomaattisten testien ongelmana on ollut niiden keskittyminen staattiseen analyysiin, jolla
\nongelmia ei pystytä löytämään tehokkaasti. Pieni määrä ihmisiä, jotka ovat katselmoineet
\nLinux-käyttöjärjestelmäytimen koodia on seurausta tehtävän vaativuudesta. Linuxin koodin
\nkatselmointi ei vaadi ainoastaan kykyä ymmärtää Linux-käyttöjärjestelmäytimen rakennetta,
\nvaan myös laitestoa tai teknologiaa johon koodimuutoksella vaikutettaan.
\nJotta nähtäisiin kuinka moderneja syväoppimistekniikoita ja malleja voitaisiin hyödyntää tilanteen
\nparantamiseksi luotiin joukko toisistaan periytyviä kehitysaskeleita. Tavoitteena ei ollut
\nainoastaan helpottaa ongelmaa koodin katselmoinnissa, vaan nähdä miten syväoppimista
\nvoidaan Linux kehityksessä hyödyntää yleisesti.
\nEnsimmäisenä tavoitteena oli luoda malli, jolla voitiin osoittaa syväoppimisen tarjoavan
\nmahdollisuuksia Linux kehityksessä. Ensimmäisen tavoitteen mallin tuli kyetä löytämään
\nperusongelmia ehdotetuista koodimuutoksista, mutta tähän mallia ei ollut mahdollista kouluttaa
\njohtuen huonolaatuisesta opetusdatasta. Oletuksena oli ollut, että jos koodinkatselmointi johti
\nuuteen versioon koodimuutoksesta, koodimuutoksen laatu paranee. Tämä oletus ei täysin
\nvastannut odotuksia.
\nTästä huolimatta, malli, joka kehitettiin tätä tehtävää varten, koulutettiin ensin ymmärtämään
\nLinuxin lähdekoodia yleisellä tasolla, ja malli selkeästi kykeni.
\nKouluttamalla malli, joka ymmärtää Linuxin lähdekoodia osoittaa, että syväoppimista voidaan
\nhyödyntää Linuxin kehitystyössä. Sama malli voidaan jatkokouluttaa moniin eri tehtäviin.
\nEpäonnistuminen kouluttaa malli löytämään virheitä koodimuutoksista ei johtunut mallin
\narkkitehtuurista, vaan opetusdatan heikosta laadusta.

Cited by 0Open access

Assimilating fission-code FIFRELIN using machine learning

Guillaume Bazelaire, A. Chebboubi, D. Bernard et al. · EPJ Web of Conferences · 2024

This paper presents work that has been done on the FIFRELIN Monte-Carlo code. The purpose of the code is to simulate the de-excitation process of fission fragments. Numerous quantity of insterest are calculated (mass yields, prompt particle spectra, mulitiplicities … ). Up to now the code relies on four free parameters which control the initial excitation and total angular momentum of fission fragment. Finding the good set of the free parameters is a diffucult task. In this work, we have developed an optimization algorithm based on Gaussian Process regression.

Cited by 0Open access

Rebuilding the Bombe [Enigma code breaking machine]

D. Lenton · IEE Review · 2001

Film makers who wanted to recreate the machines that played a vital part in breaking the German Enigma code were able to turn for advice to a group of engineers who are building their own working replica. Both film and book take their name from the coding device used by the German military to communicate battle plans. The machine was almost unbreakable in its sophistication, but procedural flaws such as reusing code book settings, made it vulnerable, and as early as 1932 Polish experts had broken the Enigma cipher and constructed a machine, the Bomba, that led to the cyphers being broken for the first time. Just before the war, they handed their work over to the British and French, and it was at Bletchley Park that mathematician Alan Turing and his colleague Gordon Welchman laid the foundations for the Bombe, electro-mechanical machine that speeded up the codebreaking process. When it came to creating several replica Bombes they commissioned South London based Asylum Models and Effects to make six machines. Asylum in turn went to a group of people who have learned the hard way the complexity of these early computing devices-a team which with support from the Bletchley Park Trust and Computer Conservation Society is rebuilding its own working Bombe. The 'Target Machine' that is under construction is based on a parts list from October 1943 for one of 68 devices built to a similar specification to decode messages from three-wheel Enigmas. This article describes the project and the history behind the building of the first Bombe machines.

Cited by 2

Python Code Smell Detection Using Machine Learning

Natthida Vatanapakorn, Chitsutha Soomlek, Pusadee Seresangtakul · 2022

Python is an increasingly popular programming language used in various software projects and domains. Code smells in Python significantly influences the maintainability, understandability, testability issues. This paper proposes a machine learning-based code smell detection for Python programs. We trained eight machine learning models with a dataset based on 115 open-source Python projects, 39 class-level software metrics, and 22 function-level software metrics. We intended to identify five code smell types in both class and function levels, i.e., long method, long parameter list, large class long scope chaining, and long based class list. Correlation-based feature selection (CFS) and logistic regression-forward stepwise (conditional) selection were employed to improve the performance of the model. This research concluded with an empirical evaluation of the performance of the machine learning approaches against the tuning machine method. The results show that the machine learning method achieved 99.72% accuracy when identifying long method and long base class list. The machine learning-based code smell detection also outperformed the tuning machine method. Moreover, we also found a set of high-impact features that contributed most when identifying each type of code smell.

Cited by 11

Code Clone Detection Using Machine Learning Techniques

Amandeep Kaur, Sandeep Sharma, Munish Saini · International Journal of Open Source Software and Processes · 2020

Code clone refers to code snippets that are copied and pasted with or without modifications. In recent years, traditional approaches for clone detection combine with other domains for better detection of a clone. This paper discusses the systematic literature review of machine learning techniques used in code clone detection. This study provides insights into various tools and techniques developed for clone detection by implementing machine learning approaches and how effectively those tools and techniques to identify clones. The authors perform a systematic literature review on studies selected from popular computer science-related digital online databases from January 2004 to January 2020. The software system and datasets used for analyzing tools and techniques are mentioned. A neural network machine learning technique is primarily used for the identification of the clone. Clone detection based on a program dependency graph must be explored in the future because it carries semantic information of code fragments.

Cited by 0

Machine Learning-Based Code-Switching and Code-Mixing Model for Bilingual Communication

Kanti Verma, Chirag Sharma · 2025

The multilingual communication use has led to evolvement of code switching and code-mixing techniques. The requirement of machine learning model is important for bilingual communication. Existing research has explored code switching in various languages, but our model integrates multiple techniques to improve accuracy. We use a Hidden Markov Model (HMM) to find changes in language in mixed-language text by looking at the data in a probabilistic way. A Support Vector Machine (SVM) classifier is utilized to detect languages at the token level. The suggested hybrid method uses a frequency-based lexicon model to look at how often words appear, a character n-gram language model to make predictions based on context, and a language-specific morphological dictionary to make classification more accurate. This integration makes it easier to sort and analyze code-switched text and solves the problems that come up when working with languages that don't have a lot of resources. Using Hidden Markov Models for sequence prediction and Support Vector Machines for feature-based classification makes it easier to work with more than one language at once. A lot of tests show that our model is 84% accurate, which is much better than traditional single-method approaches. The results indicate that integrating probabilistic, lexical, and machine learning approaches may yield more resilient code-switching models. Our research advances bilingual communication technology, facilitating enhanced natural language processing applications in multilingual communities. We also looked at how our hybrid method stacks up against top transformer-based models like mBERT and XLM-R. Transformers work very well, but they need a lot of computing power and a lot of labeled data.

Cited by 0

Learning to Generate Pseudo-Code from Source Code Using Statistical Machine Translation

Yusuke Oda, Hiroyuki Fudaba, Graham Neubig et al. · NAIST Digital Library (Nara Institute of Science and Technology) · 2015

Pseudo-code written in natural language can aid the comprehension of source code in unfamiliar programming languages. However, the great majority of source code has no corresponding pseudo-code, because pseudo-code is redundant and laborious to create. If pseudo-code could be generated automatically and instantly from given source code, we could allow for on-demand production of pseudo-code without human effort. In this paper, we propose a method to automatically generate pseudo-code from source code, specifically adopting the statistical machine translation (SMT) framework. SMT, which was originally designed to translate between two natural languages, allows us to automatically learn the relationship between source code/pseudo-code pairs, making it possible to create a pseudo-code generator with less human effort. In experiments, we generated English or Japanese pseudo-code from Python statements using SMT, and find that the generated pseudo-code is largely accurate, and aids code understanding.

Cited by 260Open access

From Machine Translation to Code-Switching: Generating High-Quality Code-Switched Text

Association for Computational Linguistics 2021, Jyothi, Preethi, Kumar, Syamantak et al. · Underline Science Inc. · 2021

Generating code-switched text is a problem of growing interest, especially given the scarcity of corpora containing large volumes of real code-switched text. In this work, we adapt a state-of-the-art neural machine translation model to generate Hindi-English code-switched sentences starting from monolingual Hindi sentences. We outline a carefully designed curriculum of pretraining steps, including the use of synthetic code-switched text, that enable the model to generate high-quality code-switched text. Using text generated from our model as data augmentation, we show significant reductions in perplexity on a language modeling task, compared to using text from other generative models of CS text. We also show improvements using our text for a downstream code-switched natural language inference task. Our generated text is further subjected to a rigorous evaluation using a human evaluation study and a range of objective metrics, where we show performance comparable (and sometimes even superior) to code-switched text obtained via crowd workers who are native Hindi speakers.

Cited by 0Open access

Specific Language + Machine Independent Machine Dependent Front End AST Intermediate Code Generator TAC Code Generator Assembly Language Sparc, x86

Anoop Sarkar, Tac -‐address Code · 2014

• Instructions that operate on named locations and labels: “generic assembly” • Locations – Every location is some place to store 4 bytes • Pretend we can make infinitely many of them – Either on stack frame: • You assign offset (plus other information possibly) – Or global variable • Referred to by global name • Labels (you generate as needed) 12-‐12-‐08 3TAC: 3-‐Address Code Addresses/Locations • names/labels: we allow source-‐program names in TAC, implemented as a pointer to a symbol table entry • constants • temporaries Instructions: • assignments: x = y op z / x = op y • copy: x = y • unconditional jump: goto L • conditional jumps: if x goto L / ifFalse x goto L / if x relop y goto L 12-‐12-‐08 <, ==,>=, etc.

Cited by 0