social network extraction from texts

... Text Extraction from Name Cards Using Neural Network Lin Lin School of Computing National University of Singapore Singapore, ... problem of text extraction from name card images with fanciful design containing various graphical foreground and reverse contrast regions. The proposed method is to apply a neural network on canny ... spatial features for the neural network and confirm that the inter-relationship of neighboring characters is crucial in distinguishing most graphical objects from texts. Table 1 shows the recall...

... prohibited. power from the network structure around them and from their position in the network. Factors that determine the influence of a social network on decision making include accessibility to network, network ... influenced by the social context in which they are embedded and by the position of actors in social networks (Granovetter, 1985). Research on social networks indicates that network structure ... processes. Social networks within an organization can reveal information related to a job, an employee, a group (team, department, unit, etc.), or the whole organization. We discuss input from social network...

... Thought and Practice: The Influence of Social Network Theory of Social Network Theory Presented at the Presented at the Networks and Complex Systems Networks and Complex Systems A talk series ... Leveraging social media to drive marketing relationships relationships – Viral marketing via the Internet Viral marketing via the Internet – Social Networking Sites Social Networking Sites ... of culture in social networks… We know very little about the role of culture in social networks… any new studies could inform us! any new studies could inform us! • Leveraging social media to...

... N e do Generate z n e,i from Multinomial(θ e ). Generate the entity n e,i from Multinomial(β z n e,i ). end for for each date which co-occurs with e, i = 1 . . . N d do Generate z d e,i from Multinomial(θ e ). Generate ... high enough to display to users (in a ranked list). In addition to being less likely to come from extraction errors, highly ranked entity/date pairs are more likely to relate to popular or important ... two key related strands of research: extracting specific types of events from Twitter, and extracting open-domain events from news [43]. Recently there has been much interest in information ex- traction...

... Wiebe. 2003. Learning Extraction Patterns for Subjective Expressions. In Proceedings of EMNLP’03. Seon Yang and Youngjoong Ko. 2009. Extracting Comparative Sentences from Korean Text Documents ... as a non-comparative sentence from a linguistic point of view. However, we conclude that this kind of sentence is as important as the other explicit comparisons from an engineering point of ... comparative sentence extraction task, we also conducted experiments for type classification using the same features (continuous POS tags sequences within a radius of 3 words from each CK) and...

... sources from existing, mature components within the translation process. This paper presents a method of phrase extraction from alignment data generated by IBM Models. By working directly from alignment ... We estimate translation con- fidence by measures from three models; the estima- tion from the maximum approximation (alignment map), estimation from the word based translation lexicon, and language ... instance from several possible transla- tions. Correctly translating function words dramati- cally boosts the NIST measure as it places emphasis on high perplexity words ie. those with diverse con- texts. 9...

... example, the candidate terms (NATIONAL NETWORK, REGIONAL NETWORK, DIS- PATCHING NETWORK) give the context (NATIONAL, REGIONAL, DISPATCHING) for the noun NETWORK. If we suppose that the modifiers ... (e.g. REGIONAL NETWORK is extracted by LEXTER), and a '0' otherwise ("national line" is not extracted by LEXTER). NATIONAL DISPATCHING REGIONAL LINE 0 1 0 NETWORK 1 1 ... obtained from the two different data sets. 4 The conceptual analysis : the LEXICLASS software LEXICLASS is a clustering tool written using C lan- guage and specialised data analysis functions from...

... inthe marketingmix.ButForrester’smostrecentupdatestotheirTechnographicsresearchsuggestthattheonlineopportunityisonly gettingbigger: “Whereisthegrowthinconsumptionofonlinecontentcoming from? From olderpeople… Social activityiswayupamong 35‐to‐44year‐olds,especiallywhenitcomestojoining social networksandreadingandreactingtocontent.Evenamong45‐ to‐54year‐olds,68%arenowSpectators,24%areJoiners,andonly28%areInactives. “Here’swhatitmeans.Itwillsoonbenomoreremarkablethatyourgrandmotherreadsablogthanthatshereadsemail. Social contentisgoingmainstream. Social contentrankshighonsearchenginesbecauseitchangessofrequentlyandgets    www.pr‐squared.com 14 Ourhumblescannerlitupthemommysphere.Duetothissinglecontestrunningon3sites,wegeneratedover80follow‐onblog postsabouttheNEATCompany’sscanner,andalmost1,200readercommentsabout“whatI’ddowithaNEATscanner.” Thisinformationwaspuregoldtoourclient–whohadnotonlyrecentlydiscoveredanewmarketopportunity,butnowalsohad accesstohundredsofpages’worthoffree,user‐generatedcontentandmarketresearch,whichtheycouldusetoinspirefuture productdevelopmentandmessagingideas. Beatspayingthousandsofdollarsforafocusgroup.  Here’sanotherexample ...  Doesthatprospectmakeyoufeelnervous?Good,itshould.Soherearesomequestionstogetyoustartedinyourinvestigation:    www.pr‐squared.com 10 Haveofficial“rulesofengagement”re:bloggersbeendocumentedanddisseminated? Haveallmarketingemployeesreportingtocorporatemarketing(and/ortoabusinessunit)beeneducatedonthe“rulesof engagement”re:externalblogs? Isthereachancethatanyofthebusinessunitmarketinggroupshaveengagedinapaidorunpaidbloggerrelationscampaign already? Ifanyofyourmarketingmanagershavepaidbloggerstowriteaboutourproducts/services,didtheyrequirepublicdisclosure oftherelationship? (Ifyoudigupunsettlinganswers,starttheprocessofdisclosinganypaidbloggerrelationshipsASAP.Youcanworryabout “whatwentwrong”later.) Bytheway,doyoutrulyalreadyhave“rulesofengagement”whenitcomestooutreachtobloggers?Thetipsonthenextpagewill getyoustarted.It’sintheformofabookmark,soyoucanprintoutthepageandtakeascissortoit.AtSHIFT,thereisalaminated versionofthisBloggerRelationsBookmarkoneveryemployee’sworkstation.          Whatabout Social MediaReleases? Idon’tsupposeIcouldwriteane‐bookabout Social MediaMarketingwithoutmentioningmy“baby,”the Social MediaRelease (SMR). Whatisa Social MediaRelease,inessence?A Social MediaReleaseconvertsthe100‐yearoldtext‐heavypressreleaseintoablog post,withallofthecoolfunctionsofablog–includingmultimedia,theabilitytocommentandaggregateinboundlinks,etc.SMRs canpoweradirectconversationwithstakeholdersaboutyournews,andgivethemthecontentelementstocarryyourmessage forwardontheirown.Cool,right? Buthere’saradicalsuggestion:don’tputany Social MediaReleasesoutoverthenewswires. WhileIsincerelyapplaudhowfarthewireservices(BusinessWire,Marketwire,PRNewswire,etal.)haveprogressedinall‐things‐ social, Iamunconvincedthat“distribution”istheBigIssuefor Social MediaReleaseadoption.Thepowerofthe Social Media Releaseisnotaboutdistribution,it’saboutempowermentandconversation.  ... it'scertainlynomorenefariousthanthetargetingtechniquesthatbigadvertisers,politicalcampaigns, etal.currentlyusetotrytoinfluencepeople. So,now,extendthisapproachwithyourownbrand. Whatifyouareamarketingexecatacompanythatsellsdigitalcameras?Howaboutcheckingoutthislinktoseethemostpopular del.icio.uspoststaggedwiththephrase“digitalcamera.”Youseethatthetoplinkistothewell‐regardedblog,DigitalPhotography Review,whichhasbeensavedbyover17,000del.icio.ususers. Maybethat’stoomany?Maybeyouwanttofocusan“edgework”campaignonthecutting‐edgeFlickrcommunity?Cool.The#2 most‐savedlinkusingthe“digitalcamera”tagisthisone:“Top10camerasonFlickr–Digitalcameranews– Logicamera.com,”whichhasbeensavedby111people.Whatcanyoudowiththisinformation…? Example:Perhapsyourclientisabigcameracompany,whichwantsto createafocusgroupforanupcoming“dummy‐proof”dSLRbyrecruiting peoplewhohavepubliclysavedatleastten“digitalphotography101–style” links. Firstoff,youcouldcreateamicrositecontainingaspecialoffertopeople whovolunteerforthisfocusgroup,andthen“saveitfor”thisnewbiegroup whomyou’veidentifiedviathedel.icio.ustagsearchstrategywe’veoutlinedhere. Onehundredandninety‐fourpeoplehavesavedthisarticleabout“WhichlensesformyDigitalSLR?”Giventhe“tutorial”toneof thearticle,it’ssafetoassumethatthese90peoplearelikelygettingseriousaboutphotographybutmaystillbefairlynewtothe burgeoningdSLRfield.Youcouldcross‐referencethese90usersagainstthe115userswhosavedthe“Top10camerasonFlickr”link mentionedaboveandvoila:you’veidentifiedexactlythetypesofpeoplewhomaybehappytoserveinafocusgroupaboutyour client’snewdSLRcamera–especiallyifyouofferthemabigcouponforyourbrandedgoods,ormaybeconsidergiftingthemwitha FlickrProaccount.Inreturnyou’dgetterrificfeedback from savvyonlineconsumersaboutyourproductandbrand…andpossibly thebeginningsofabuzzcampaignaboutyourupcomingproductlaunch. Whenyou’relivingontheedge,thepossibilitiesareendless.    Ihopethecontentinthise‐bookhelpedyouout.I’dlovetohearyourfeedback!Ifyou’vegotquestionsorcomments,dropmea lineattdefren@shiftcomm.com.Andbesuretosubscribetomyblog,www.pr‐squared.comformoreinsights,ideasandramblings. Lastbutnotleast,ifyou’reamarketerwhocoulduseouragency’shelp,hoponovertowww.shiftcomm.com. Ifyouthinkbacktosomeofthedown‐and‐dirtyideasdiscussedinthise‐book,you’llrealizethattheseconceptsofparticipationand findabilityareinextricablylinkedtothetactics.Forexample,byreachingoutdirectlyto social bookmarkusers,youareparticipating inacommunityandensuringthatyourcontentbecomesmorefindable–throughyourdirectactionwithusers,aswellasby ensuringthatmorepeoplesaveandshareyourcontent(withtheaffiliatedtagsthatbenefityourbrandandyourSEO).Subtlebut importantstuff,eh? linkedtomoreoften,somoreandmoreonlineadultsarebecomingexposedtoit,acceptingit,andembracingit.Ifyou’rea marketer,nomatterwhatgroupofconsumersyou’retargeting,thismeansyoumustpayattentiontothe social world online.” www.pr‐squared.com 40 ### Whatagreattimetobeamarketer! Now,talkamongstyourselves!    www.pr‐squared.com 34 Forexample:TheBabyBoomersarenowreachingretirementage.ThatmeansthattheWWIIgenerationisnowenteringits twilight.AndthatmeansthatplentyofBoomersaregoingtobecaringfortheirownchildrenaswellastheirelderlyparents,with increasingfrequency.AccordingtotheAmericanAssociationofHomesandServicesfortheAging,between2007and2015the numberofAmericansaged85andolderisexpectedtoincreaseby40percent…andby2020,12millionolderAmericanswillneed long‐termhealthcare. Thepeoplemostaffectedbythesetrendsarealreadydesperateforsolaceandcounsel,and theirnumberswillonlygrow.But,mostareSpectators.Wherecantheygo?Thethriving communityatasitelikeEonsistinycomparedtothelargerneed;besides,asnoted,Spectators don’tevenjoin social networks. Whiletherearesomegoodresourcesavailableonline,virtuallyallarecurrentlystatic,“1.0”sites thatwillonlybefoundviaatypicalGooglesearch.Evenwheremessageboardsareavailable, theytendtobescattershot,redundant,and/orthe“authority”ofthemoderatorsis questionable. Meanwhile,theblogosphereappearstobeveryquietontheseissues.(Why?Cuzthefolks facingthesechallengesareSpectators!Theyarelookingforcontent,notcreatingit.) Thereisamajoropportunityforalargehealthcareprovider,insuranceagency,ornursing homesyndicatetocreatecompelling,interactive,ongoingcontentthataddressesthisneed. Imaginethethoughtleadershippotentialforabrandthatgetsthisright,justasthechallengeis bearingdownonanentiregeneration? Thatdoesn’tmean“createanew social network. ”Yetagain:theSpectatorsdon’tjoin social networks! SohowdoyoureachtheSpectators?     www.pr‐squared.com 4 Ofcoursetherearechallengesforanycompany–evenofBMW’ssize–to scaleandtrainandmonitoragroupof“communitymanagers”thatcould serveasadjunctstothemarketinggroup.AlsoIacknowledgethatthiswasan unscientificapproach.ForallIknow,BMWisengaging“big‐time”inother partsoftheblogosphere. BMW’s“failure”inthisexperiment,wasnottakingadvantageofaneasily‐ identifiedopportunitytoengage,consistently,inaplacethatissoprominently impactingtheirbrand. Ifyou’renotpayingattentiontowhat’sbeingsaidaboutyourbrandwithintheTop10searchresults,howcanyouclaimtobe “engaged?” Googleiseverybrand’snewhomepage.JustaskSpiritAirlines:the#5organicsearchresultatthetimeofthiswriting?Ablogpost thatsays“DoNotFlySpiritAirlines,”whichwassopoorlyhandledbythecompanythatitspawnedallkindsoftrouble. ParticipationisMarketing.Thebettertheparticipation–intermsoftone,responsivenessandfrequency–thebetterthemarketing. Thinkaboutwhatthecustomerswanttohear,notwhatyouwantthemtohear.(Andcertainlynotwhatyouwantthemtobuy. Social Mediaisnottherightforumforasalespitch.Ifyouworkonbuildingloyaltytoyourbrand’spersonality,saleswillfollowin duecourse.) So,if“participation”isthegoal,howdoyougetstarted?     www.pr‐squared.com 31 Remember,youonlyget255characterspaceswithdel.icio.us,so,keepitshort&sweet,likeso: “Noticedyou’dsavedSHIFT’s Social MediaNewsReleasetemplate.Thoughtyoumightalsobeinterestedinthis Social Media Newsroomtemplate.There’saPPTavailable,whichdescribesthewholeapproach,atshiftcomm.com.I’m attodd@shiftcomm.com.” That’s247characterspaces,inwhichwe’veexplainedHOWwefoundtheperson(relevancy),WHYweare“intruding,”andWHERE theycangetmoreinfoifinterested,includinghow‐towriteback,formoreinfo(ortokvetch). It’simportanttonotethatmanypeoplewhousedel.icio.usforpersonalreasonsdraftquicknotes(tothemselvesand/ortothe community)aboutWHYtheysavedthelink.Thisallowsforamoremeticulousapproach,i.e.,ifsomeonehasnotedthattheydislike the Social MediaNewsReleasetemplate,theyshouldn’tgetthegenericnoteusedabove!Wecouldignorethem,ortryasofter approach. Nowwhat? Oncetheoutreachhasbeenmadeviathe“for:username”function,somefolksonthereceivingendofourcampaignwillelectto “accept”thislinkbyactivelysavingitwithintheiraccount(ametricyoucantrack).Somefolkswilltakefurtheraction:theymight blogaboutit(measurable);theymightdownloadthetemplate(measurable);theymightcommentattheoriginalNewsroom’sdebut blogpost(measurable).Theymightemailyoudirectly(measurable).Theymightdonothing(alsomeasurable). Inanyeventwecanfeelprettyconfidentthatwe’veONLYreachedouttopeoplewhowouldlikelybegenuinelyinterested,basedon theirprior,public, social bookmarkingbehavior.Ifwe’vekickstartedtheconversationabout Social MediaNewsroomsa‐fresh,it’s likelygoingtobeasmoothandproductivetransitionwiththiscrowd. Harkeningbacktomystatementthat“yourmileagemayvary,”youshouldkeepinmindthatnotallend‐userswillappreciatebeing reachedouttointhisway.ButtheapproachIamdiscussinghereallowsthemarketertonarrowtheiroutreachbasedon identifiable,contextualand(Idaresay)legitimateapproaches:“YoupubliclysavedTHIS;youevenpubliclysaidWHYyousavedit...

... ability to model so- cial power lects as well. Apart from text classification, work from the topic modeling community is also closely related to Social Power Modeling. Andrew McCallum ex- tended ... greatly aid analysts doing national intelligence work. Social network analysis (SNA) presupposes a collection of individuals, whereas a social power lect classifier, once trained, would provide ... extend to social power modeling, and our approach may well be applicable to other di- mensions of social meaning. In the coming sections, we first establish the Related Work, primarily from Statistical...

... of 60- 70% (Huang et al., 2000). The task that is most similar to our work is named entity extraction from speech data (DARPA, 1999). Although the goal of the named entity task is similar - to ... stochastic- transducer induction. It aims to learn rules auto- matically from training data instead of requiring hand-crafted rules from experts. Although the re- sults with this system are not yet ... voicemail mes- sages. duces the number of features from to with minor performance loss. This shows that the main power of the maxent model comes from a a very small subset of the possible features....

... the progressive reduction of the size of training corpora: e.g., from the 1,000 texts of the MUC-5 (MUC-5, 1993) to the 100 texts in MUC-6 (MUC-6, 1995). When the cor- pus size is limited, ... 227 Proceedings of EACL '99 The Development of Lexical Resources for Information Extraction from Text Combining WordNet and Dewey Decimal Classification* Gabriela Cavagli~t ITC-irst ... hypernyms and some coordinated terms. The proposed methodology is corpus centered (starting from the corpus analysis to build the Core Lexicon) and can always be profitably ap- plied. It...

... frames. The network society, in the simplest terms, is a social structure based on networks operated by information and communication tech- nologies based in microelectronics and digital computer networks ... networks. Naturally, these programs are decided socially from outside the net- work. But once they are inscripted in the logic of the network, the network will follow efficiently these instructions, ... knowledge recognized as such by the scientific community) has found in various social contexts. The Network Society: From Knowledge to Policy 7 K85232_01.qxp 12/27/05 1:37 PM Page 7 The communication...

... and grey hair, Mr. Merriman was a man whose appearance from the first claimed interest. It was a few days after his Budget speech, which, from various innovations, had aroused a storm of criticism, ... iron, imported, I was told, from Wolverhampton. This roofing, indeed, prevails over the whole of new South Africa; and although it appears a very unsuitable protection from the burning rays of ... officials sent out from the mother-country of the Dutch. They looked on the Transvaal only as a means for getting rich. Hence the fearful state of bribery and corruption among them, from the highest...

... parallel content extraction from comparable corpora. It consists of tools bundled in two workflows: (1) alignment of comparable documents and extraction of parallel sentences and (2) extraction ... parallel sentence pairs are extracted from the aligned comparable corpora (section 2.2). The workflow for named entity (NE) and terminology extraction and mapping from comparable corpora extracts ... LEXACC requires aligned document pairs (also m to n alignments) for sentence extraction. It also allows extraction from comparable corpora as a whole; however, precision may decrease due to...

... content within a certain period and/or from a certain group of peo- ple such as people in the same region. Existing work on keyphrase extraction identifies keyphrases from either individual documents ... range from 0.45 to 0.80, showing fair to good agreement 2 . We further dis- carded all candidates with an average score less than 1. The number of the remaining keyphrases for each topic ranges from ... keyphrase extraction is evaluated using precision and recall on all the extracted keyphrases. We choose not to use these measures for the fol- lowing reasons: (1) Traditional keyphrase extraction works...