Open-Access-Bücher zur Sprachwissenschaft

In der let­zten Zeit sind u.a. diese frei ver­füg­baren Titel erschienen:

The Origins of Self: An Anthropological Perspective

Mar­tin P. J. Edwardes |

The Ori­gins of Self explores the role that self­hood plays in defin­ing human soci­ety, and each human indi­vid­ual in that soci­ety. It con­sid­ers the genet­ic and cul­tur­al ori­gins of self, the role that self plays in social­i­sa­tion and lan­guage, and the types of self we gen­er­ate in our indi­vid­ual jour­neys to and through adult­hood.

Edwardes argues that oth­er aware­ness is a rel­a­tive­ly ear­ly evo­lu­tion­ary devel­op­ment, present through­out the pri­mate clade and per­haps beyond, but self-aware­ness is a prod­uct of the shar­ing of social mod­els, some­thing only humans appear to do. The self of which we are aware is not some­thing innate with­in us, it is a mod­el of our self pro­duced as a response to the mod­els of us offered to us by oth­er peo­ple. Edwarde­spro­pos­es that human con­struc­tion of self­hood involves sev­en dif­fer­ent types of self. All but one of them are inter­nal­ly gen­er­at­ed mod­els, and the only non-mod­el, the actu­al self, is com­plete­ly hid­den from con­scious aware­ness. We rely on oth­ers to tell us about our self, and even to let us know we are a self.Developed in rela­tion to a range of sub­ject areas – lin­guis­tics, anthro­pol­o­gy, genomics and cog­ni­tion, as well as socio-cul­tur­al the­o­ry – The Ori­gins of Self is of par­tic­u­lar inter­est to stu­dents and researchers study­ing the ori­gins of lan­guage, human ori­gins in gen­er­al, and the cog­ni­tive dif­fer­ences between human and oth­er ani­mal psy­cholo­gies.

Corpus linguistics: A guide to the methodology

Ana­tol Ste­fanow­itsch |

Cor­po­ra are wide­ly used in lin­guis­tics, but not always wise­ly. This book attempts to frame cor­pus lin­guis­tics sys­tem­at­i­cal­ly as a vari­ant of the obser­va­tion­al method. The first part intro­duces the read­er to the gen­er­al method­olog­i­cal dis­cus­sions sur­round­ing cor­pus data as well as the prac­tice of doing cor­pus lin­guis­tics, includ­ing issues such as the sci­en­tif­ic research cycle, research design, extrac­tion of cor­pus data and sta­tis­ti­cal eval­u­a­tion. The sec­ond part con­sists of a num­ber of case stud­ies from the main areas of cor­pus lin­guis­tics (lex­i­cal asso­ci­a­tions, mor­phol­o­gy, gram­mar, text and metaphor), sur­vey­ing the range of issues stud­ied in cor­pus lin­guis­tics while at the same time show­ing how they fit into the method­ol­o­gy out­lined in the first part.

Linguistics for the age of AI

Mar­jorie McShane & Sergei Niren­burg |

One of the orig­i­nal goals of arti­fi­cial intel­li­gence research was to endow intel­li­gent agents with human-lev­el nat­ur­al lan­guage capa­bil­i­ties. Recent AI research, how­ev­er, has focused on apply­ing sta­tis­ti­cal and machine learn­ing approach­es to big data rather than attempt­ing to mod­el what peo­ple do and how they do it. In this book, Mar­jorie McShane and Sergei Niren­burg return to the orig­i­nal goal of recre­at­ing human-lev­el intel­li­gence in a machine. They present a human-inspired, lin­guis­ti­cal­ly sophis­ti­cat­ed mod­el of lan­guage under­stand­ing for intel­li­gent agent sys­tems that empha­sizes meaning—the deep, con­text-sen­si­tive mean­ing that a per­son derives from spo­ken or writ­ten lan­guage.

With Lin­guis­tics for the Age of AI, McShane and Niren­burg offer a roadmap for cre­at­ing lan­guage-endowed intel­li­gent agents (LEIAs) that can understand,explain, and learn. They describe the lan­guage-under­stand­ing capa­bil­i­ties of LEIAs from the per­spec­tives of cog­ni­tive mod­el­ing and sys­tem build­ing, empha­siz­ing “actionability”—which involves achiev­ing inter­pre­ta­tions that are suf­fi­cient­ly deep, pre­cise, and con­fi­dent to sup­port rea­son­ing about action. After detail­ing their microthe­o­ries for top­ics such as seman­tic analy­sis, basic coref­er­ence, and sit­u­a­tion­al rea­son­ing, McShane and Niren­burg turn to agent appli­ca­tions devel­oped using those microthe­o­ries and eval­u­a­tions of a LEIA’s lan­guage under­stand­ing capa­bil­i­ties.

McShane and Niren­burg argue that the only way to achieve human-lev­el lan­guage under­stand­ing by machines is to place lin­guis­tics front and cen­ter, using sta­tis­tics and big data as con­tribut­ing resources. They lay out a long-term research pro­gram that address­es lin­guis­tics and real-world rea­son­ing togeth­er, with­in a com­pre­hen­sive cog­ni­tive archi­tec­ture.

Schreibe einen Kommentar

Pflichtfelder sind mit * markiert.