{"id":"https://openalex.org/W1502614524","doi":null,"title":"Smoothing a probablistic lexicon via syntactic transformations","display_name":"Smoothing a probablistic lexicon via syntactic transformations","publication_year":2001,"publication_date":"2001-01-01","ids":{"openalex":"https://openalex.org/W1502614524","mag":"1502614524"},"language":"en","primary_location":{"id":"pmh:oai:repository.upenn.edu:dissertations-2818","doi":null,"is_oa":false,"landing_page_url":"https://repository.upenn.edu/dissertations/AAI3015308","pdf_url":null,"source":{"id":"https://openalex.org/S4377196331","display_name":"Scholarly Commons (University of Pennsylvania)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I79576946","host_organization_name":"University of Pennsylvania","host_organization_lineage":["https://openalex.org/I79576946"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Dissertations available from ProQuest","raw_type":"text"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5052467896","display_name":"Jason M. Eisner","orcid":"https://orcid.org/0000-0002-8861-0772","observed_orcids":["https://orcid.org/0000-0002-8861-0772"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jason Eisner","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5111565465","display_name":"Mitchell P. Marcus","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mitchell P. Marcus","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":16,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9984999895095825,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9984999895095825,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10034","display_name":"Syntax, Semantics, Linguistic Variation","score":0.00039999998989515007,"subfield":{"id":"https://openalex.org/subfields/1203","display_name":"Language and Linguistics"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.0003000000142492354,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/probabilistic-parsing","display_name":"probabilistic parsing","score":0.8040000200271606},{"id":"https://openalex.org/keywords/syntactic-transformation","display_name":"syntactic transformation","score":0.7289999723434448},{"id":"https://openalex.org/keywords/meta-rules","display_name":"meta-rules","score":0.9100000262260437},{"id":"https://openalex.org/keywords/transformation-rate","display_name":"transformation rate","score":0.6940000057220459},{"id":"https://openalex.org/keywords/bayesian-smoothing","display_name":"Bayesian smoothing","score":0.6470000147819519},{"id":"https://openalex.org/keywords/log-linear-models","display_name":"log-linear models","score":0.9879999756813049},{"id":"https://openalex.org/keywords/lexical-entries","display_name":"lexical entries","score":0.3659999966621399},{"id":"https://openalex.org/keywords/perplexity-reduction","display_name":"perplexity reduction","score":0.5789999961853027},{"id":"https://openalex.org/keywords/model-averaging","display_name":"model averaging","score":0.828000009059906},{"id":"https://openalex.org/keywords/syntactic-preferences","display_name":"syntactic preferences","score":0.3479999899864197}],"concepts":[{"id":"https://openalex.org/C2778121359","wikidata":"https://www.wikidata.org/wiki/Q8096","display_name":"Lexicon","level":2,"score":0.5853085517883301},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5014853477478027},{"id":"https://openalex.org/C186644900","wikidata":"https://www.wikidata.org/wiki/Q194152","display_name":"Parsing","level":2,"score":0.4887516498565674},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.4647485613822937},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4321981966495514},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.42548519372940063},{"id":"https://openalex.org/C204241405","wikidata":"https://www.wikidata.org/wiki/Q461499","display_name":"Transformation (genetics)","level":3,"score":0.42395472526550293},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.34362494945526123},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:repository.upenn.edu:dissertations-2818","doi":null,"is_oa":false,"landing_page_url":"https://repository.upenn.edu/dissertations/AAI3015308","pdf_url":null,"source":{"id":"https://openalex.org/S4377196331","display_name":"Scholarly Commons (University of Pennsylvania)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I79576946","host_organization_name":"University of Pennsylvania","host_organization_lineage":["https://openalex.org/I79576946"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Dissertations available from ProQuest","raw_type":"text"},{"id":"mag:1502614524","doi":null,"is_oa":false,"landing_page_url":"https://www.cs.jhu.edu/~jason/papers/eisner.thesis01-submitted.pdf","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":194,"referenced_works":["https://openalex.org/W18940505","https://openalex.org/W39478293","https://openalex.org/W79411081","https://openalex.org/W86030598","https://openalex.org/W173367458","https://openalex.org/W174472240","https://openalex.org/W1408813051","https://openalex.org/W1489958605","https://openalex.org/W1491161564","https://openalex.org/W1491175367","https://openalex.org/W1493034779","https://openalex.org/W1496659931","https://openalex.org/W1504046386","https://openalex.org/W1507020546","https://openalex.org/W1512277306","https://openalex.org/W1514127556","https://openalex.org/W1515383272","https://openalex.org/W1520976986","https://openalex.org/W1525914480","https://openalex.org/W1527142120","https://openalex.org/W1528941926","https://openalex.org/W1530250655","https://openalex.org/W1531010694","https://openalex.org/W1535015163","https://openalex.org/W1537988943","https://openalex.org/W1539312853","https://openalex.org/W1542522816","https://openalex.org/W1543028224","https://openalex.org/W1548013757","https://openalex.org/W1550951390","https://openalex.org/W1551104980","https://openalex.org/W1559171143","https://openalex.org/W1560263223","https://openalex.org/W1562275107","https://openalex.org/W1567570606","https://openalex.org/W1568426241","https://openalex.org/W1572100238","https://openalex.org/W1575431606","https://openalex.org/W1575798196","https://openalex.org/W1576269293","https://openalex.org/W1581740421","https://openalex.org/W1582364809","https://openalex.org/W1582714098","https://openalex.org/W1585559843","https://openalex.org/W1586060904","https://openalex.org/W1586368951","https://openalex.org/W1588039225","https://openalex.org/W1591380337","https://openalex.org/W1592796124","https://openalex.org/W1597533204","https://openalex.org/W1598510668","https://openalex.org/W1600844763","https://openalex.org/W1601728146","https://openalex.org/W1615454278","https://openalex.org/W1622422412","https://openalex.org/W1632114991","https://openalex.org/W1647671624","https://openalex.org/W1658676592","https://openalex.org/W1674799117","https://openalex.org/W1809415035","https://openalex.org/W1825362480","https://openalex.org/W1843392930","https://openalex.org/W1859173823","https://openalex.org/W1966307998","https://openalex.org/W1966553486","https://openalex.org/W1966812932","https://openalex.org/W1972573551","https://openalex.org/W1973759805","https://openalex.org/W1975674805","https://openalex.org/W1977472488","https://openalex.org/W1977547765","https://openalex.org/W1978470410","https://openalex.org/W1981724541","https://openalex.org/W1982944197","https://openalex.org/W1986543644","https://openalex.org/W1991923024","https://openalex.org/W1996672843","https://openalex.org/W1997895564","https://openalex.org/W2007780422","https://openalex.org/W2008172256","https://openalex.org/W2012730307","https://openalex.org/W2016589492","https://openalex.org/W2016789247","https://openalex.org/W2020755048","https://openalex.org/W2024184126","https://openalex.org/W2024657257","https://openalex.org/W2029184000","https://openalex.org/W2038248725","https://openalex.org/W2039217078","https://openalex.org/W2040713190","https://openalex.org/W2040985480","https://openalex.org/W2047706513","https://openalex.org/W2048679005","https://openalex.org/W2049633694","https://openalex.org/W2052449326","https://openalex.org/W2056573988","https://openalex.org/W2064070185","https://openalex.org/W2064318101","https://openalex.org/W2065203287","https://openalex.org/W2066636486","https://openalex.org/W2067551521","https://openalex.org/W2071521298","https://openalex.org/W2072435322","https://openalex.org/W2073272771","https://openalex.org/W2079656678","https://openalex.org/W2088563966","https://openalex.org/W2095958485","https://openalex.org/W2096175520","https://openalex.org/W2096466920","https://openalex.org/W2096945176","https://openalex.org/W2098379588","https://openalex.org/W2098750402","https://openalex.org/W2101210369","https://openalex.org/W2101711363","https://openalex.org/W2102381086","https://openalex.org/W2102997946","https://openalex.org/W2104637942","https://openalex.org/W2105934661","https://openalex.org/W2107126982","https://openalex.org/W2107671947","https://openalex.org/W2108227097","https://openalex.org/W2109601663","https://openalex.org/W2111907245","https://openalex.org/W2115389092","https://openalex.org/W2116316001","https://openalex.org/W2118097783","https://openalex.org/W2118119027","https://openalex.org/W2118382442","https://openalex.org/W2118782751","https://openalex.org/W2119610041","https://openalex.org/W2119794571","https://openalex.org/W2121647878","https://openalex.org/W2122978558","https://openalex.org/W2124556666","https://openalex.org/W2124591829","https://openalex.org/W2125529971","https://openalex.org/W2127314673","https://openalex.org/W2128084896","https://openalex.org/W2130630493","https://openalex.org/W2130747650","https://openalex.org/W2130851608","https://openalex.org/W2131058166","https://openalex.org/W2131103009","https://openalex.org/W2131814656","https://openalex.org/W2132807457","https://openalex.org/W2132957691","https://openalex.org/W2134199742","https://openalex.org/W2134237567","https://openalex.org/W2134666210","https://openalex.org/W2135625884","https://openalex.org/W2135976285","https://openalex.org/W2136028896","https://openalex.org/W2136834963","https://openalex.org/W2138787466","https://openalex.org/W2142708806","https://openalex.org/W2142843952","https://openalex.org/W2145269499","https://openalex.org/W2147880316","https://openalex.org/W2152810530","https://openalex.org/W2157477135","https://openalex.org/W2158195707","https://openalex.org/W2158312965","https://openalex.org/W2158388102","https://openalex.org/W2159080219","https://openalex.org/W2160783091","https://openalex.org/W2160842254","https://openalex.org/W2160943512","https://openalex.org/W2162792036","https://openalex.org/W2164141546","https://openalex.org/W2166451556","https://openalex.org/W2169528473","https://openalex.org/W2170521549","https://openalex.org/W2170716495","https://openalex.org/W2205493714","https://openalex.org/W2328580899","https://openalex.org/W2338181011","https://openalex.org/W2344991153","https://openalex.org/W2422872931","https://openalex.org/W2480595210","https://openalex.org/W2567948266","https://openalex.org/W2571532437","https://openalex.org/W2616709338","https://openalex.org/W2752885492","https://openalex.org/W2766736793","https://openalex.org/W2913184320","https://openalex.org/W2950268498","https://openalex.org/W2952654140","https://openalex.org/W2981264952","https://openalex.org/W3044664353","https://openalex.org/W3088178427","https://openalex.org/W3089160278","https://openalex.org/W3089319657","https://openalex.org/W3121926921","https://openalex.org/W3140677416"],"related_works":["https://openalex.org/W7052767019","https://openalex.org/W2084921315","https://openalex.org/W2106434447","https://openalex.org/W2101733803","https://openalex.org/W2520421163","https://openalex.org/W2074260075","https://openalex.org/W2090079220","https://openalex.org/W7100557381","https://openalex.org/W2760572670","https://openalex.org/W1909398668"],"abstract_inverted_index":{"Probabilistic":[0],"parsing":[1,279],"requires":[2],"a":[3,60,64,85,97,108,123,176,187,224,286,325],"lexicon":[4],"that":[5,21,46,169,227],"specifies":[6],"each":[7],"word's":[8],"syntactic":[9],"preferences":[10],"in":[11,104],"terms":[12],"of":[13,32,81,87,96,115,295,299],"probabilities.":[14,220],"To":[15],"estimate":[16,78],"these":[17],"probabilities":[18,114],"for":[19],"words":[20],"were":[22],"poorly":[23],"observed":[24,313],"during":[25],"training,":[26],"this":[27,300],"thesis":[28],"assumes":[29],"the":[30,53,79,82,105,113,131,146,153,162,183,214,262,269,277,293,307],"existence":[31],"arbitrarily":[33],"powerful":[34],"transformations":[35,68,83],"(also":[36],"known":[37],"to":[38,77,149,192,197,205,273,291,304,311,320],"linguists":[39],"as":[40],"lexical":[41,61,88,116,124],"redundancy":[42],"rules":[43],"or":[44,51,100,137,245],"metarules)":[45],"can":[47,111,238],"add,":[48],"delete,":[49],"retype":[50],"reorder":[52],"argument":[54],"and":[55,71,186,252,284,315],"adjunct":[56],"positions":[57],"specified":[58],"by":[59,213,242,246],"entry.":[62],"In":[63],"given":[65],"language,":[66],"some":[67,316],"apply":[69],"frequently":[70],"others":[72],"rarely.":[73],"We":[74,267],"describe":[75],"how":[76,151],"rates":[80,148],"from":[84,158,171,276,306,317],"sample":[86],"entries.":[89,117,160],"More":[90],"deeply,":[91],"we":[92,110],"learn":[93],"which":[94],"properties":[95],"transformation":[98,308],"increase":[99],"decrease":[101],"its":[102,318],"rate":[103],"language.":[106],"As":[107],"result,":[109],"smooth":[112],"Given":[118],"enough":[119],"direct":[120],"evidence":[121,136,139],"about":[122],"entry's":[125],"probability,":[126],"our":[127],"Bayesian":[128],"approach":[129],"trusts":[130],"evidence;":[132],"but":[133],"when":[134],"less":[135],"no":[138],"is":[140,189,302],"available,":[141],"it":[142],"relies":[143],"more":[144],"on":[145,261],"transformations'":[147],"guess":[150],"often":[152],"entry":[154],"will":[155],"be":[156,239],"derived":[157],"related":[159],"Abstractly,":[161],"proposed":[163],"\u201ctransformation":[164],"models\u201d":[165],"are":[166,203,256,264],"probability":[167],"distributions":[168],"arise":[170],"graph":[172],"random":[173,194],"walks":[174,195],"with":[175,258],"log-linear":[177],"parameterization.":[178],"A":[179],"domain":[180,207],"expert":[181],"constructs":[182],"parameterized":[184],"graph,":[185],"vertex":[188],"likely":[190],"according":[191],"whether":[193],"tend":[196],"halt":[198],"at":[199],"it.":[200],"Transformation":[201],"models":[202,222],"suited":[204],"any":[206],"where":[208],"\u201crelated\u201d":[209],"events":[210],"(as":[211],"defined":[212],"graph)":[215],"may":[216],"have":[217],"positively":[218],"covarying":[219],"Such":[221],"admit":[223],"natural":[225],"prior":[226],"favors":[228],"simple":[229],"regular":[230],"relationships":[231],"over":[232],"stipulative":[233],"exceptions.":[234],"The":[235],"model":[236],"parameters":[237],"locally":[240],"optimized":[241],"gradient-based":[243],"methods":[244],"Expectation-Maximization.":[247],"Exact":[248],"algorithms":[249],"(matrix":[250],"inversion)":[251],"approximate":[253],"ones":[254],"(relaxation)":[255],"provided,":[257],"optimizations.":[259],"Variations":[260],"idea":[263],"also":[265],"discussed.":[266],"compare":[268],"new":[270],"technique":[271],"empirically":[272],"previous":[274],"techniques":[275],"probabilistic":[278],"literature,":[280],"using":[281],"comparable":[282],"features,":[283],"obtain":[285],"20%":[287],"perplexity":[288,328],"reduction":[289,301],"(similar":[290],"doubling":[292],"amount":[294],"training":[296],"data).":[297],"Some":[298],"shown":[303],"stem":[305],"model's":[309],"ability":[310,319],"match":[312],"probabilities,":[314],"generalize.":[321],"Model":[322],"averaging":[323],"yields":[324],"final":[326],"24%":[327],"reduction.":[329]},"counts_by_year":[{"year":2015,"cited_by_count":1},{"year":2012,"cited_by_count":1}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2016-06-24T00:00:00"}