{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2023-07-07 17:13:01.457105: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n" ] } ], "source": [ "import pandas as pd\n", "import gensim\n", "import pprint\n", "from gensim import corpora\n", "from gensim.utils import simple_preprocess\n", "from gensim.models import TfidfModel\n", "from gensim.parsing import strip_tags, strip_numeric, \\\n", " strip_multiple_whitespaces, stem_text, strip_punctuation, \\\n", " remove_stopwords, preprocess_string\n", "import re\n", "import os\n", "\n", "from typing import List\n", "import spacy" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "transform_to_lower = lambda s: s.lower()\n", "remove_single_char = lambda s: re.sub(r'\\s+\\w{1}\\s+', '', s)\n", "\n", "cleaning_filters = [\n", " strip_tags,\n", " strip_numeric,\n", " strip_punctuation, \n", " strip_multiple_whitespaces, \n", " transform_to_lower,\n", " remove_stopwords,\n", " remove_single_char\n", "]" ] }, { "cell_type": "code", "execution_count": 95, "metadata": {}, "outputs": [], "source": [ "df = pd.read_parquet(\"/Users/luis.morales/Desktop/arxiv-paper-recommender/data/processed/reduced_arxiv_papers.parquet.gzip\")" ] }, { "cell_type": "code", "execution_count": 94, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "638707" ] }, "execution_count": 94, "metadata": {}, "output_type": "execute_result" } ], "source": [ "int(df.shape[0] * 0.75) " ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 73, "metadata": {}, "outputs": [], "source": [ "df = pd.read_parquet(\"/Users/luis.morales/Desktop/arxiv-paper-recommender/data/processed/reduced_arxiv_papers.parquet.gzip\").sample().reset_index(drop=True)" ] }, { "cell_type": "code", "execution_count": 74, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
idsubmitterauthorstitlecommentsjournal-refdoireport-nocategorieslicenseabstractversionsupdate_dateauthors_parsedcleaned_abstractslen_abstract
02007.00905Song QinghengQingheng Song, Yong Zeng, Jie Xu, and Shi JinA Survey of Prototype and Experiment for UAV C...24 pages, 6 figuresNoneNoneNonecs.IT eess.SP math.IThttp://creativecommons.org/licenses/by-nc-sa/4.0/Unmanned aerial vehicle (UAV) communications...[{'created': 'Thu, 2 Jul 2020 06:26:20 GMT', '...2020-07-03[[Song, Qingheng, ], [Zeng, Yong, ], [Xu, Jie,...unmanned aerial vehicle uav communication a...865
12102.04209Michael StuartMichael T. Stuart and Markus KneerGuilty Artificial Minds20 pages, 4 figures, 1 tableNoneNoneNonecs.CY cs.AI cs.HChttp://creativecommons.org/licenses/by/4.0/The concepts of blameworthiness and wrongnes...[{'created': 'Sun, 24 Jan 2021 21:37:35 GMT', ...2021-02-09[[Stuart, Michael T., ], [Kneer, Markus, ]]concept blameworthiness wrongness fundament...739
21201.5796Denis JeromeDenis JeromeOrganic Superconductors: when correlations and...41 pages, 21 figures to be published in Journa...None10.1007/s10948-012-1475-7Nonecond-mat.supr-conhttp://arxiv.org/licenses/nonexclusive-distrib...This survey provides a brief account for the...[{'created': 'Fri, 27 Jan 2012 15:24:46 GMT', ...2012-02-21[[Jerome, Denis, ]]survey provide brief account start organic ...649
31511.03076Emma Platts MissGeorge F.R. Ellis, Emma Platts, David Sloan an...Current observations with a decaying cosmologi...23 pages, 11 figuresNone10.1088/1475-7516/2016/04/026Noneastro-ph.CO gr-qc hep-thhttp://arxiv.org/licenses/nonexclusive-distrib...We use the phase plane analysis technique of...[{'created': 'Tue, 10 Nov 2015 12:08:23 GMT', ...2016-04-27[[Ellis, George F. R., ], [Platts, Emma, ], [S...use phase plane analysis technique madsen e...554
41710.02954Kirk BansakKirk BansakEstimating Causal Moderation Effects with Rand...Forthcoming, Journal of the Royal Statistical ...NoneNoneNonestat.MEhttp://arxiv.org/licenses/nonexclusive-distrib...Researchers are often interested in analyzin...[{'created': 'Mon, 9 Oct 2017 06:34:01 GMT', '...2020-08-25[[Bansak, Kirk, ]]researcher interested analyze conditional t...799
...................................................
8516051301.0707Sebastian KleinSebastian KleinChow groups of tensor triangulated categories40 pages. The presentation of the article has ...NoneNoneNonemath.AG math.CT math.RThttp://arxiv.org/licenses/nonexclusive-distrib...We recall P. Balmer's definition of tensor t...[{'created': 'Fri, 4 Jan 2013 11:06:40 GMT', '...2015-10-02[[Klein, Sebastian, ]]recall p. balmer definition tensor triangul...787
8516061707.00341Giorgos AnastasiouGiorgos Anastasiou, Rodrigo Olea, David Rivera...Noether-Wald energy in Critical Gravity7 pages, no figures, Final version for PLBNone10.1016/j.physletb.2018.11.021Nonehep-th gr-qchttp://arxiv.org/licenses/nonexclusive-distrib...Criticality represents a specific point in t...[{'created': 'Sun, 2 Jul 2017 19:52:32 GMT', '...2018-11-21[[Anastasiou, Giorgos, ], [Olea, Rodrigo, ], [...criticality represent specific point parame...631
8516071610.08526Blagoje OblakBlagoje OblakBMS Particles in Three Dimensions437 pages (including index), 33 figures. Appen...None10.1007/978-3-319-61878-4Nonehep-th gr-qc math-ph math.GR math.MP math.RThttp://arxiv.org/licenses/nonexclusive-distrib...This thesis is devoted to the group-theoreti...[{'created': 'Wed, 26 Oct 2016 20:00:16 GMT', ...2018-01-29[[Oblak, Blagoje, ]]thesis devoted group theoretic aspect dimen...542
8516081211.6629Philippe JoyezPhilippe JoyezSelf-consistent dynamics of a Josephson juncti...7 pages, 1 figureNone10.1103/PhysRevLett.110.217003Nonecond-mat.supr-con cond-mat.mes-hallhttp://arxiv.org/licenses/nonexclusive-distrib...We derive microscopically the dynamics assoc...[{'created': 'Tue, 27 Nov 2012 17:29:04 GMT', ...2013-05-29[[Joyez, Philippe, ]]derive microscopically dynamic associate d....558
8516090705.2878Benoit PerthameBenoit Perthame (DMA), Panagiotis E. SouganidisAsymmetric potentials and motor effect: a larg...NoneNoneNoneNonemath.APNoneWe provide a mathematical analysis of appear...[{'created': 'Sun, 20 May 2007 17:43:39 GMT', ...2007-05-23[[Perthame, Benoit, , DMA], [Souganidis, Panag...provide mathematical analysis appearance co...518
\n", "

851610 rows × 16 columns

\n", "
" ], "text/plain": [ " id submitter \n", "0 2007.00905 Song Qingheng \\\n", "1 2102.04209 Michael Stuart \n", "2 1201.5796 Denis Jerome \n", "3 1511.03076 Emma Platts Miss \n", "4 1710.02954 Kirk Bansak \n", "... ... ... \n", "851605 1301.0707 Sebastian Klein \n", "851606 1707.00341 Giorgos Anastasiou \n", "851607 1610.08526 Blagoje Oblak \n", "851608 1211.6629 Philippe Joyez \n", "851609 0705.2878 Benoit Perthame \n", "\n", " authors \n", "0 Qingheng Song, Yong Zeng, Jie Xu, and Shi Jin \\\n", "1 Michael T. Stuart and Markus Kneer \n", "2 Denis Jerome \n", "3 George F.R. Ellis, Emma Platts, David Sloan an... \n", "4 Kirk Bansak \n", "... ... \n", "851605 Sebastian Klein \n", "851606 Giorgos Anastasiou, Rodrigo Olea, David Rivera... \n", "851607 Blagoje Oblak \n", "851608 Philippe Joyez \n", "851609 Benoit Perthame (DMA), Panagiotis E. Souganidis \n", "\n", " title \n", "0 A Survey of Prototype and Experiment for UAV C... \\\n", "1 Guilty Artificial Minds \n", "2 Organic Superconductors: when correlations and... \n", "3 Current observations with a decaying cosmologi... \n", "4 Estimating Causal Moderation Effects with Rand... \n", "... ... \n", "851605 Chow groups of tensor triangulated categories \n", "851606 Noether-Wald energy in Critical Gravity \n", "851607 BMS Particles in Three Dimensions \n", "851608 Self-consistent dynamics of a Josephson juncti... \n", "851609 Asymmetric potentials and motor effect: a larg... \n", "\n", " comments journal-ref \n", "0 24 pages, 6 figures None \\\n", "1 20 pages, 4 figures, 1 table None \n", "2 41 pages, 21 figures to be published in Journa... None \n", "3 23 pages, 11 figures None \n", "4 Forthcoming, Journal of the Royal Statistical ... None \n", "... ... ... \n", "851605 40 pages. The presentation of the article has ... None \n", "851606 7 pages, no figures, Final version for PLB None \n", "851607 437 pages (including index), 33 figures. Appen... None \n", "851608 7 pages, 1 figure None \n", "851609 None None \n", "\n", " doi report-no \n", "0 None None \\\n", "1 None None \n", "2 10.1007/s10948-012-1475-7 None \n", "3 10.1088/1475-7516/2016/04/026 None \n", "4 None None \n", "... ... ... \n", "851605 None None \n", "851606 10.1016/j.physletb.2018.11.021 None \n", "851607 10.1007/978-3-319-61878-4 None \n", "851608 10.1103/PhysRevLett.110.217003 None \n", "851609 None None \n", "\n", " categories \n", "0 cs.IT eess.SP math.IT \\\n", "1 cs.CY cs.AI cs.HC \n", "2 cond-mat.supr-con \n", "3 astro-ph.CO gr-qc hep-th \n", "4 stat.ME \n", "... ... \n", "851605 math.AG math.CT math.RT \n", "851606 hep-th gr-qc \n", "851607 hep-th gr-qc math-ph math.GR math.MP math.RT \n", "851608 cond-mat.supr-con cond-mat.mes-hall \n", "851609 math.AP \n", "\n", " license \n", "0 http://creativecommons.org/licenses/by-nc-sa/4.0/ \\\n", "1 http://creativecommons.org/licenses/by/4.0/ \n", "2 http://arxiv.org/licenses/nonexclusive-distrib... \n", "3 http://arxiv.org/licenses/nonexclusive-distrib... \n", "4 http://arxiv.org/licenses/nonexclusive-distrib... \n", "... ... \n", "851605 http://arxiv.org/licenses/nonexclusive-distrib... \n", "851606 http://arxiv.org/licenses/nonexclusive-distrib... \n", "851607 http://arxiv.org/licenses/nonexclusive-distrib... \n", "851608 http://arxiv.org/licenses/nonexclusive-distrib... \n", "851609 None \n", "\n", " abstract \n", "0 Unmanned aerial vehicle (UAV) communications... \\\n", "1 The concepts of blameworthiness and wrongnes... \n", "2 This survey provides a brief account for the... \n", "3 We use the phase plane analysis technique of... \n", "4 Researchers are often interested in analyzin... \n", "... ... \n", "851605 We recall P. Balmer's definition of tensor t... \n", "851606 Criticality represents a specific point in t... \n", "851607 This thesis is devoted to the group-theoreti... \n", "851608 We derive microscopically the dynamics assoc... \n", "851609 We provide a mathematical analysis of appear... \n", "\n", " versions update_date \n", "0 [{'created': 'Thu, 2 Jul 2020 06:26:20 GMT', '... 2020-07-03 \\\n", "1 [{'created': 'Sun, 24 Jan 2021 21:37:35 GMT', ... 2021-02-09 \n", "2 [{'created': 'Fri, 27 Jan 2012 15:24:46 GMT', ... 2012-02-21 \n", "3 [{'created': 'Tue, 10 Nov 2015 12:08:23 GMT', ... 2016-04-27 \n", "4 [{'created': 'Mon, 9 Oct 2017 06:34:01 GMT', '... 2020-08-25 \n", "... ... ... \n", "851605 [{'created': 'Fri, 4 Jan 2013 11:06:40 GMT', '... 2015-10-02 \n", "851606 [{'created': 'Sun, 2 Jul 2017 19:52:32 GMT', '... 2018-11-21 \n", "851607 [{'created': 'Wed, 26 Oct 2016 20:00:16 GMT', ... 2018-01-29 \n", "851608 [{'created': 'Tue, 27 Nov 2012 17:29:04 GMT', ... 2013-05-29 \n", "851609 [{'created': 'Sun, 20 May 2007 17:43:39 GMT', ... 2007-05-23 \n", "\n", " authors_parsed \n", "0 [[Song, Qingheng, ], [Zeng, Yong, ], [Xu, Jie,... \\\n", "1 [[Stuart, Michael T., ], [Kneer, Markus, ]] \n", "2 [[Jerome, Denis, ]] \n", "3 [[Ellis, George F. R., ], [Platts, Emma, ], [S... \n", "4 [[Bansak, Kirk, ]] \n", "... ... \n", "851605 [[Klein, Sebastian, ]] \n", "851606 [[Anastasiou, Giorgos, ], [Olea, Rodrigo, ], [... \n", "851607 [[Oblak, Blagoje, ]] \n", "851608 [[Joyez, Philippe, ]] \n", "851609 [[Perthame, Benoit, , DMA], [Souganidis, Panag... \n", "\n", " cleaned_abstracts len_abstract \n", "0 unmanned aerial vehicle uav communication a... 865 \n", "1 concept blameworthiness wrongness fundament... 739 \n", "2 survey provide brief account start organic ... 649 \n", "3 use phase plane analysis technique madsen e... 554 \n", "4 researcher interested analyze conditional t... 799 \n", "... ... ... \n", "851605 recall p. balmer definition tensor triangul... 787 \n", "851606 criticality represent specific point parame... 631 \n", "851607 thesis devoted group theoretic aspect dimen... 542 \n", "851608 derive microscopically dynamic associate d.... 558 \n", "851609 provide mathematical analysis appearance co... 518 \n", "\n", "[851610 rows x 16 columns]" ] }, "execution_count": 74, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df" ] }, { "cell_type": "code", "execution_count": 75, "metadata": {}, "outputs": [], "source": [ "corpus = df['cleaned_abstracts'].to_list()\n" ] }, { "cell_type": "code", "execution_count": 76, "metadata": {}, "outputs": [], "source": [ "def gensim_tokenizer(docs: List[str]):\n", " tokenized_docs = list()\n", " for doc in docs:\n", " processed_words = preprocess_string(doc, cleaning_filters)\n", " tokenized_docs.append(processed_words)\n", " \n", " return tokenized_docs\n", "\n", "tokenized_corpus = gensim_tokenizer(corpus)" ] }, { "cell_type": "code", "execution_count": 77, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "851610" ] }, "execution_count": 77, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(tokenized_corpus)" ] }, { "cell_type": "code", "execution_count": 55, "metadata": {}, "outputs": [], "source": [ "def cleaning_pipe(document):\n", " # Invoking gensim.parsing.preprocess_string method with set of filters\n", " processed_words = preprocess_string(document, cleaning_filters)\n", " return processed_words" ] }, { "cell_type": "code", "execution_count": 78, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "/Users/luis.morales/Desktop/arxiv-paper-recommender/models\n", "/Users/luis.morales/Desktop/arxiv-paper-recommender\n" ] }, { "data": { "text/plain": [ "False" ] }, "execution_count": 78, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def validate_if_dictionary_exists(dictionaty_name: str) -> bool:\n", " dicts_folder = \"models/nlp_dictionaries\"\n", " current_dir = os.getcwd()\n", " parent_dir = os.path.dirname(current_dir)\n", " dict_path = f\"{parent_dir}/{dicts_folder}/{dictionaty_name}\"\n", " print(current_dir)\n", " print(parent_dir)\n", " return os.path.isfile(dict_path)\n", " \n", "\n", "validate_if_dictionary_exists('30ktokens.dict') " ] }, { "cell_type": "code", "execution_count": 79, "metadata": {}, "outputs": [], "source": [ "def get_gensim_dictionary(tokenized_docs: List[str], dict_name: str = \"corpus\", save_dict: bool = False):\n", " \"\"\"\n", " Create dictionary of words in preprocessed corpus and saves the dict object\n", " \"\"\"\n", " dictionary = corpora.Dictionary(tokenized_docs)\n", " if save_dict: \n", " dict_lenght = len(tokenized_corpus)\n", " parent_folder = \"/Users/luis.morales/Desktop/arxiv-paper-recommender/models/dictionaries\"\n", " #if validate_if_dictionary_exists('30ktokens.dict'):\n", " dictionary.save(f'{parent_folder}/{dict_name}.dict')\n", " \n", " return dictionary\n", "\n", "dictionary = get_gensim_dictionary(tokenized_docs=tokenized_corpus, dict_name=\"TextualTango\", save_dict=True)" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [], "source": [ "# def get_gensim_dictionary(tokenized_docs: List[str], dict_name: str = \"corpus\", save_dict: bool = False):\n", "# \"\"\"\n", "# Create dictionary of words in preprocessed corpus and saves the dict object\n", "# \"\"\"\n", "# dictionary = corpora.Dictionary(tokenized_docs)\n", "# if save_dict: \n", "# dict_lenght = len(tokenized_corpus)\n", "# parent_folder = \"/Users/luis.morales/Desktop/arxiv-paper-recommender/models/nlp_dictionaries\"\n", "# if validate_if_dictionary_exists('30ktokens.dict'):\n", "# dictionary.save(f'{parent_folder}/{dict_name}.dict')\n", " \n", "# return dictionary\n", "\n", "# dictionary = get_gensim_dictionary(tokenized_docs=tokenized_corpus, dict_name=\"300Ktokens\", save_dict=True)" ] }, { "cell_type": "code", "execution_count": 80, "metadata": {}, "outputs": [], "source": [ "BoW_corpus = [dictionary.doc2bow(doc, allow_update=True) for doc in tokenized_corpus]" ] }, { "cell_type": "code", "execution_count": 81, "metadata": {}, "outputs": [], "source": [ "tfidf_model = TfidfModel(BoW_corpus)" ] }, { "cell_type": "code", "execution_count": 82, "metadata": {}, "outputs": [], "source": [ "tfidf_model.save(\"/Users/luis.morales/Desktop/arxiv-paper-recommender/models/tfidf/TextualTango.model\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Test model" ] }, { "cell_type": "code", "execution_count": 83, "metadata": {}, "outputs": [], "source": [ "# index the tfidf vector of corpus as sparse matrix\n", "from gensim import similarities\n", "index = similarities.SparseMatrixSimilarity(tfidf_model[BoW_corpus], num_features=len(dictionary))" ] }, { "cell_type": "code", "execution_count": 84, "metadata": {}, "outputs": [], "source": [ "index.save(\"/Users/luis.morales/Desktop/arxiv-paper-recommender/models/similarities_matrix/TextualTangoSimilarities/TextualTango\")" ] }, { "cell_type": "code", "execution_count": 67, "metadata": {}, "outputs": [], "source": [ "def get_closest_n(query, n):\n", " '''get the top matching docs as per cosine similarity\n", " between tfidf vector of query and all docs'''\n", " query_document = cleaning_pipe(query)\n", " query_bow = dictionary.doc2bow(query_document)\n", " sims = index[tfidf_model[query_bow]]\n", " top_idx = sims.argsort()[-1*n:][::-1]\n", " return top_idx" ] }, { "cell_type": "code", "execution_count": 85, "metadata": {}, "outputs": [], "source": [ "def get_recomendations_metadata(query: str, n: int, df: pd.DataFrame):\n", " recommendations_idxs = get_closest_n(query, n)\n", " return df.iloc[recommendations_idxs].reset_index(drop=True)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 91, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "User Request ---- : \n", " Which papers discuss the use of statistical models and Bayesian inference for uncertainty quantification and risk assessment in engineering systems?\n", "User Request ---- : \n", " \n", "Title: A framework for benchmarking uncertainty in deep regression\n", "Abstract: We propose a framework for the assessment of uncertainty quantification in\n", "deep regression. The framework is based on regression problems where the\n", "regression function is a linear combination of nonlinear functions. Basically,\n", "any level of complexity can be realized through the choice of the nonlinear\n", "functions and the dimensionality of their domain. Results of an uncertainty\n", "quantification for deep regression are compared against those obtained by a\n", "statistical reference method. The reference method utilizes knowledge of the\n", "underlying nonlinear functions and is based on a Bayesian linear regression\n", "using a reference prior. Reliability of uncertainty quantification is assessed\n", "in terms of coverage probabilities, and accuracy through the size of calculated\n", "uncertainties. We illustrate the proposed framework by applying it to current\n", "approaches for uncertainty quantification in deep regression. The flexibility,\n", "together with the availability of a reference solution, makes the framework\n", "suitable for defining benchmark sets for uncertainty quantification.\n", "\n", "\n", "--------------------------\n", "User Request ---- : \n", " Which papers discuss the use of statistical models and Bayesian inference for uncertainty quantification and risk assessment in engineering systems?\n", "User Request ---- : \n", " \n", "Title: Generative Parameter Sampler For Scalable Uncertainty Quantification\n", "Abstract: Uncertainty quantification has been a core of the statistical machine\n", "learning, but its computational bottleneck has been a serious challenge for\n", "both Bayesians and frequentists. We propose a model-based framework in\n", "quantifying uncertainty, called predictive-matching Generative Parameter\n", "Sampler (GPS). This procedure considers an Uncertainty Quantification (UQ)\n", "distribution on the targeted parameter, which matches the corresponding\n", "predictive distribution to the observed data. This framework adopts a\n", "hierarchical modeling perspective such that each observation is modeled by an\n", "individual parameter. This individual parameterization permits the resulting\n", "inference to be computationally scalable and robust to outliers. Our approach\n", "is illustrated for linear models, Poisson processes, and deep neural networks\n", "for classification. The results show that the GPS is successful in providing\n", "uncertainty quantification as well as additional flexibility beyond what is\n", "allowed by classical statistical procedures under the postulated statistical\n", "models.\n", "\n", "\n", "--------------------------\n", "User Request ---- : \n", " Which papers discuss the use of statistical models and Bayesian inference for uncertainty quantification and risk assessment in engineering systems?\n", "User Request ---- : \n", " \n", "Title: Recent Advances in Uncertainty Quantification Methods for Engineering\n", " Problems\n", "Abstract: In the last few decades, uncertainty quantification (UQ) methods have been\n", "used widely to ensure the robustness of engineering designs. This chapter aims\n", "to detail recent advances in popular uncertainty quantification methods used in\n", "engineering applications. This chapter describes the two most popular\n", "meta-modeling methods for uncertainty quantification suitable for engineering\n", "applications (Polynomial Chaos Method and Gaussian Process). Further, the UQ\n", "methods are applied to an engineering test problem under multiple\n", "uncertainties. The test problem considered here is a supersonic nozzle under\n", "operational uncertainties. For the deterministic solution, an open-source\n", "computational fluid dynamics (CFD) solver SU2 is used. The UQ methods are\n", "developed in Matlab and are further combined with SU2 for the uncertainty and\n", "sensitivity estimates. The results are presented in terms of the mean and\n", "standard deviation of the output quantities.\n", "\n", "\n", "--------------------------\n" ] } ], "source": [ "_input = \"Which papers discuss the use of statistical models and Bayesian inference for uncertainty quantification and risk assessment in engineering systems?\"\n", "results_df = get_recomendations_metadata(query=_input, df=df, n=3)\n", "\n", "\n", "for abstract in list(zip(results_df['abstract'].to_list(), results_df['title'].to_list())):\n", " print(f\"User Request ---- : \\n {_input}\")\n", " print(f\"User Request ---- : \\n \")\n", " print(f\"Title: {abstract[1]}\")\n", " print(f\"Abstract: {abstract[0]}\\n\")\n", " print(f\"--------------------------\")" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3.11.4 ('arxiv-env': venv)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.11.4" }, "orig_nbformat": 4, "vscode": { "interpreter": { "hash": "aae17c2ae2f38cc6f211be9b71a2aa280701d8462782cbc1f67caa83a1603363" } } }, "nbformat": 4, "nbformat_minor": 2 }