JP4487332B2 - Information processing apparatus and method, recording medium, and information processing system - Google Patents

Description

式（２）の第２項の分母Ｍは、共有体験項目数を表す。この例の場合、共有体験項目数Ｍは４となる。式（２）の第２項の分子は、各共有体験項目ＴＰに対する注目会員と比較対象会員の好感度の差異を表す好感差異ｅiの総和であり、好感差異ｅiは、式（３）に基づいて求められる。
【００８９】
ｅi＝｜ＫAi−ＫBi｜×ＷA（ＫAi）×ＷB（ＫBi）・・・（３）
ＫAiおよびＫBi（なお、以下においては、ＫAiとＫBiを特に区別する必要がない場合、両者をまとめて、単にＫiと記述する）は、それぞれ、共有体験項目ＴＰに対する注目会員および比較対象会員の好感度である。すなわち、式（３）の右辺の第１因子である｜ＫAi−ＫBi｜により、共有体験項目ＴＰに対する注目会員の好感度ＫAｉと比較対象会員の好感度ＫBｉの差が求められる。
【００９０】
式（３）の右辺の第２因子であるＷA（ＫAi）および第３因子であるＷB（ＫBi）（なお、以下においては、ＷA（ＫAi）とＷB（ＫBi）を特に区別する必要がない場合、両者をまとめて、単にＷ（Ｋi）と記述する）は、それぞれ、注目会員と比較対象会員の共有体験項目ＴＰに対する好感度Ｋiの曖昧さに対応する加重関数であり、式（４）に基づいて算出される。なお、式（４）において、０≦α≦１である。
【００９１】
Ｗ（Ｋi）＝４α（Ｋi−０．５)２＋（１−α）・・・（４）
好感度Ｋiが1.0の場合、または好感度Ｋiが0.0の場合、Ｗ（Ｋi）は最大値1.0となり、また、好感度Ｋiが0.5の場合、Ｗ（Ｋi）は最小値（１−α）となる。
【００９２】
好感度Ｋiが0.5とされた場合、その共有体験項目ＴＰに対しての好感は、最も良い好感と最も悪い好感の中間の好感であり、良くも悪くもないという最も曖昧さを含む評価がなされたものと考えられる。そこで、好感度Ｋiが0.5に近い値であるほど、すなわち、より曖昧さを含む評価がなされるほど、Ｗ（Ｋi）の値が小さくなるようにし、その結果、好感差異ｅiの値が小さくなるようにし、もって、最終的に求められる好感類似度Ｐ２に評価の曖昧さが含まれることを抑制する。
【００９３】
なお、上述した注目会員に対応する加重関数ＷA（ＫAi）と比較対象会員に対応する加重関数ＷB（ＫBi）において、異なるα値を設定したり、それぞれ異なる加重関数に基づいて、評価の曖昧さに対応する重みを算出することもできる。また、α値を０とすることにより、曖昧さに対応する重みを考慮しない好感類似度Ｐ２を算出することもできる。
【００９４】
なお、好感差異は、式（３）を計算する代わりに、例えば、図２２に示すようなルックアップテーブルを参照することにより算出することもできる。このルックアップテーブルでは、横軸に好感度ＫAiが、また、縦軸に好感度ＫBiが、それぞれ示されており、２つの値により特定される値が好感差異として求められる。例えば、横軸の好感度ＫAiが0.3であり、また、縦軸の好感度ＫBiが0.8である場合、好感差異は0.4となる。
【００９５】
図２１に戻り、ステップＳ２０４において、CPU２１は、ステップＳ２０３で算出した好感類似度Ｐ２が、予め設定してある閾値より大きいか否かを判定し、算出した好感類似度Ｐ２が閾値より大きい場合、ステップＳ２０５に進み、その好感類似度Ｐ２を、同好度Ｆとする。なお、ステップＳ２０１で求められた体験共有率Ｐ１と、ステップＳ２０３で算出され好感類似度Ｐ２の内分した値を同好度Ｆとするなど、体験共有率Ｐ１により好感類似度Ｐ２を調整して、同好度Ｆを求めることもできる。
【００９６】
ステップＳ２０２において、ステップＳ２０１で算出された体験共有率Ｐ１が閾値より小さいと判定された場合、または、ステップＳ２０４において、ステップ２０３で算出された好感類似度Ｐ２が、閾値より小さいと判定された場合、ステップＳ２０６に進み、CPU２１は、同好度を０とする。すなわち、この場合、比較対象会員の嗜好と注目会員の嗜好とは、類似していないものとされる。
【００９７】
このように、図２１に示した同好度Ｆの算出処理が終了すると、次に図１７のステップＳ１０６に進む。
【００９８】
なお、以上においては、体験共有率Ｐ１は、式（１）に基づいて算出されたが式（５）に基づいて算出することもできる。
【００９９】
体験共有率Ｐ１＝（共有体験項目数Ｍ）／（注目会員と比較対象会員の少なくとも一方が体験した体験項目数総数Ｎmab） ・・・（５）
この例の場合、注目会員（ユーザA）が体験した体験項目数総数は７であり（図１８）、比較対象会員（ユーザＢ）が体験した体験項目数総数は９である（図２０）が、そのうちの、４つの体験項目（CDタイトル２０１乃至２０４）が共有体験項目であるので、注目会員と比較対象会員の少なくとも一方が体験した体験項目数総数Ｎmabは１２（＝７＋９−４）となり、体験共有率Ｐ１は４／１２となる。すなわち、式（１）により求められる体験共有率Ｐ１は、比較対象会員の体験項目数総数に関係なく求められるが、式（５）により求められる体験共有率Ｐ１は、比較対象会員の体験項目総数により変化する。
【０１００】
また、以上においては、好感類似度Ｐ２は、式（２）に基づいて算出したが、式（６）に基づいて算出することもできる。
【０１０１】
【数２】

式（６）の第２項のうち、式（７）で表される分子の値は、共有体験項目ＴＰに対して注目会員が評価した好感度ＫAiと、比較対象会員が評価した好感度ＫBiが近い値になればなるほど小さい値になる。一方、式（６）の第２項の分母のうち、式（８）で表される第２項の値は、両者の好感度ＫAiとＫBiが近い値になり、さらに好感度ＫAiとＫBiが1.0または0.0に近い値になればなるほど大きい値になる。すなわち、式（６）に基づいて算出される好感類似度Ｐ２は、好感度ＫAiとＫBiの両方が同時に0.0または1.0に近い値である場合、すなわち、より明確な同じ結論（曖昧でない結論）がなされた場合、交換類似度Ｐ２が大きな値になるように重み付けされている。
【０１０２】
【数３】

【数４】

なお、共分散を利用して好感類似度Ｐ２を算出することもできる。この方法は、正の相関（第１または第３象限の固有ベクトル）に相当する固有値と、負の相関（第２または第４象限の固有ベクトル）に相当する固有値から好感類似度Ｐ２を算出するもので、はじめに、両者の好感度の平均値、両者の好感度Ｋｉの２乗和および積の和から共分散係数行列が求められ、その行列を対角化することにより、２つの固有値とそれに対応する固有ベクトル（主方向）が求められる。例えば、求められた２つの固有値の比をとり、１から差し引いた値が好感類似度Ｐ２とされる。
【０１０３】
図１７に戻り、ステップＳ１０６において、CPU２１は、RAM２３の中の同好者バッファ２３Ｂに、最大記憶件数Ｎfmax（図１９、図２３および図２４の例の場合、最大記憶件数ＮfmaxはＬに等しい）分の同好者データが記憶されているか否かを判定し、最大記憶件数Ｎfmax分の同好者データが記憶されている場合、ステップＳ１０７に進み、ステップＳ１０５で算出した同好度Ｆより小さい同好度Ｆを含む同好者データが同好者バッファ２３Ｂに存在するか否かを判定し、そのような同好者データが存在する場合、ステップＳ１０８に進む。
【０１０４】
ステップＳ１０８において、CPU２１は、ステップＳ１０７で確認した、ステップＳ１０５で算出した同好度Ｆより小さい同好度Ｆが含まれる同好度データのうち、同好度Ｆの値が最も小さい１つの同好度データを削除する。具体的には、図２３に示されるように、同好者バッファ２３Ｂにおいては、同好度Ｆの値が大きい同好者データが、小さい番号となるように（図中、上に位置するように）保持されている。従って、ここでは、最も番号の大きい（図２３において、番号Ｌの）同好者データが削除される。なお、ハードディスク２７の記憶容量の範囲で、同好者バッファ２３Ｂの最大記憶件数Ｎfmaxを変更することもできる。
【０１０５】
次に、ステップＳ１０９において、新しい同好者データを同好者バッファ２３Ｂに追加する処理が実行される。すなわち、CPU２１は、ステップＳ１０５で算出した同好度Ｆを含む同好者データを、同好度Ｆが高い同好者データが小さい番号に対応するように、同好者バッファ２３Ｂに記憶させる。具体的には、CPU２１は、ステップＳ１０５で算出した同好度Ｆを、最も大きい番号の同好者データの同好度Ｆと比較し、前者の方が大きければ、１つ小さい番号の同好者データの同好度Ｆと比較する処理を繰り返す。そして、CPU２１は、ステップＳ１０５で算出した同好度Ｆの値が、同好者バッファ２３Ｂに記憶されている同好者データの同好度Ｆの値と等しいか、または小さければ、その記憶されている同好者データの番号より１つ大きい番号の位置に、新たな同好者データを記憶させる。
【０１０６】
ステップＳ１１０において、CPU２１は、ハードディスク２７に、注目会員との同好度Ｆがまだ算出されていない会員（まだ、比較対象会員とされていない会員）のデータが存在するか否かを判定し、そのような会員のデータが存在する場合、ステップＳ１１１に進み、まだ、比較対象会員とされていない会員のデータの中から、次の比較対象会員の会員IDを選択し、RAM２３の比較対象会員IDバッファ２３Ｃに設定させる。その後、処理はステップＳ１０４に戻り、それ以降の処理が繰り返し実行される。
【０１０７】
CPU２１は、ステップＳ１０６において、同好者バッファ２３Ｂに最大記憶件数Ｎfmax分の同好者データが記憶されていないと判定した場合（図１９または図２４に示すように、同好者バッファ２３Ｂに記憶されている同好者データの数が、Ｌ個より小さい場合）、ステップＳ１０９に進み、ステップＳ１０５で算出した同好度Ｆを含む同好者データを、同好度Ｆの大きさの順番に従って、同好者バッファ２３Ｂに記憶させる。
【０１０８】
ステップＳ１０７において、ステップＳ１０５で算出した同好度Ｆより小さい値の同好度Ｆが存在しないと判定された場合、CPU２１は、ステップＳ１０８，Ｓ１０９の処理をスキップし（その同好者データを同好者バッファ２３Ｂに追加せず）、ステップＳ１１０に進み、注目会員との同好度Ｆが求められていない会員が存在するか否かを判定する。
【０１０９】
ステップＳ１１０において、注目会員との同好度Ｆがまだ算出されていない会員は存在しないと判定された場合、処理は終了される。
【０１１０】
このように、注目会員（ユーザA）と同様の嗜好を有する同好者が抽出され、同好者データが同好者バッファ２３Ｂに記憶される。
【０１１１】
なお、ステップＳ１０４の処理において、比較対象会員の個人体験好感レコードＲに記憶されている体験好感要素データＥを、種々の条件に基づいて選別して比較対象会員体験好感リスト２３Ｄに設定することができる。例えば、１ヶ月以内に投稿された体験好感要素データＥのみが、比較対象会員体験好感リスト２３Ｄに設定されるようにすることができ、このようにすれば、比較的新しい体験好感要素データＥに基づいて同好者を抽出することができる。この場合、この条件を満たす体験好感要素データＥを保持しない会員は、比較対象会員にはならない。また、体験好感要素データＥを選別する条件として、日時の他、体験場所など様々な条件を設定することができ、さらに複数の条件を複合することができる。
【０１１２】
また、同好者を抽出する条件として、上述した同好度Ｆ以外に、性別、年齢、住所、特定の団体に所属するか否か、あるいは電子メールなどで個人的に連絡することが可能か否かなどの条件をさらに追加することができる。実際には、CPU２１は、ステップＳ１０３において、設定された同好者抽出条件に基づいて、比較対象会員を決定する。この場合、比較対象会員の候補とされる会員に対して、各単純条件命題に対して求められた論理値から、複合条件命題に対応する理論式に従って、合成値を算出してこれを判定値にしたり、一度に複合条件命題に対する理論値を算出し、それを判定値とすることもできる。このように算出された判定値により、比較対象会員が選別されるので、同好者抽出条件を満たさない会員が同好者として抽出されることはない。
【０１１３】
また、以上においては、注目会員体験好感リスト２３Ａは、繰り返し参照されるため、RAM２３に記憶されたが、例えば、図１８に示すような構成（体験項目と好感度の欄を有する構成）を有する連想メモリ（contents addressable memory）に、注目会員体験好感リスト２３Ａを記憶させておくことで、より高速に処理を実行することができる。例えば、注目会員体験好感リスト２３Ａに記憶されるデータのうち、体験項目に設定されるデータを固定長データとして、それを連想メモリ上のアドレスデータとする。そしてそれに対応する好感度をその連想データとして、連想メモリに記憶させる。
【０１１４】
このようにすることで、例えば、ステップＳ２０１において、体験共有率Ｐ１が算出される場合、注目会員体験好感リスト２３ＡがRAM２３に記憶されているとき、比較対象会員体験好感リスト２３Ｄと注目会員体験好感リスト２３Ａの両方の内容が確認される。すなわち、この例の場合、注目会員体験好感リスト２３Ａの長さである７と、比較対象会員体験好感リスト２３Ｄの長さである９の積である６３回分の比較処理が実行される。これに対して、注目会員体験好感リスト２３Ａが連想メモリに記憶されている場合、比較対象会員好感リスト２３Ｄの長さである９回分の比較処理が行われる。
【０１１５】
次に、図１６のステップＳ７４における合成体験好感リスト生成処理の詳細を、図２５のフローチャートを参照して説明する。
【０１１６】
はじめに、ステップＳ３０１において、CPU２１は、RAM２３の中の体験好感累積バッファ２３Ｅ（図２７を参照して後述する）を初期化し、次に、ステップＳ３０２において、同好者バッファ２３Ｄに記憶されている同好者データの数をカウントするカウンタの値ｊを１に初期設定する。
【０１１７】
CPU２１は、ステップＳ３０３において、カウンタの値ｊに対応する番号の同好者データに含まれる会員IDの会員を合成対象会員とし、その会員の個人体験好感レコードＲを、ハードディスク２７から読み出し、RAM２３の合成対象会員体験好感リスト２３Ｆ（図示せず）に記憶させる。なお、ｊ＝１とされたとき、同好者バッファ２３Ｂに記憶されている同好者データのうち、先頭に位置する（番号が最も小さい）、すなわち、最も大きい同好度Ｆに対応する会員IDの会員が、合成対象会員とされる。
【０１１８】
次に、ステップＳ３０４において、CPU２１は、合成対象会員体験好感リスト２３Ｆに記憶されている体験好感要素データＥに基づいて、体験好感累積バッファ２３Ｅにデータを追加する処理を行う。
【０１１９】
ステップＳ３０４における体験好感累積バッファ２３Ｅへのデータ追加処理を、図２６のフローチャートを参照して説明する。
【０１２０】
ステップＳ４０１において、CPU２１は、合成対象会員体験好感リスト２３Ｆに記憶されている、体験項目数Ｎｉをカウントするカウンタの値ｉを１に初期設定する。
【０１２１】
次に、ステップＳ４０２において、CPU２１は、カウンタの値ｉに対応する体験好感要素データＥの体験項目Ｔiが、体験好感累積バッファ２３Ｅの体験項目リストに存在するか否かを判定し、存在すると判定した場合、ステップＳ４０３に進み、カウンタの値ｉに対応する体験好感要素データＥの好感度Ｋを、その体験項目Ｔi（CDタイトル）に対応する好感度配列２３Ｇ（図２７）に追加させる。
【０１２２】
CPU２１は、ステップＳ４０２において、カウンタの値ｉに対応する体験好感要素データＥの体験項目Ｔi（合成対象会員体験好感リスト２３Ｆに記憶されている）が、体験好感累積バッファ２３Ｅの体験項目リストに存在しないと判定した場合、ステップＳ４０４において、その体験好感要素データＥの体験項目Ｔiを、体験好感累積バッファ２３Ｅの体験項目リストの最後に設定し、ステップＳ４０５において、体験項目Ｔiに対応する好感度Ｋを、体験項目Ｔiに対応した好感度配列２３Ｇに設定させる。
【０１２３】
ステップＳ４０３またはステップＳ４０５において、好感度配列２３Ｇに好感度Ｋが設定されると、ステップＳ４０６に進み、CPU２１は、カウンタの値ｉが合成対象会員体験好感リスト２３Ｆの体験項目数Ｎと等しいか否かを判定し、両者が等しくない場合、ステップＳ４０７に進み、カウンタの値ｉを１だけインクリメントして、ステップＳ４０２に戻り、それ以降の処理を繰り返し実行する。
【０１２４】
CPU２１は、ステップＳ４０６で、カウンタの値ｉが合成対象会員体験好感リスト２３Ｆの体験項目数Ｎと等しいと判定した場合、処理を終了し、図２５のステップ３０５に進む。
【０１２５】
ステップＳ３０５において、CPU２１は、同好者バッファ２３Ｂに記憶されている同好者データの数をカウントするカウンタの値ｊが、同好者バッファ２３Ｂの最後の番号Ｌと等しいか否かを判定し、カウンタの値ｊが同好者バッファ２３Ｂの最後の番号Ｌと等しくないと判定した場合、ステップＳ３０６に進み、カウンタの値ｊを１だけインクリメントして、ステップＳ３０３に戻り、それ以降の処理を繰り返し実行する。
【０１２６】
このようにして、体験好感累積バッファ２３Ｅには、図２７に示すように、体験項目Ｔと、それに対応する好感度配列２３Ｇに好感度Ｋが設定される。
【０１２７】
CPU２１は、ステップＳ３０５で、カウンタの値ｊが同好者バッファ２３Ｂに記憶されている同好者データの数Ｌと等しいと判定した場合、ステップＳ３０７に進む。
【０１２８】
ステップＳ３０７において、CPU２１は、合成体験好感リスト作成処理を実行する。すなわち、CPU ２１は、体験好感累積バッファ２３Ｅの体験項目リストに設定された体験項目Ｔごとに、対応した好感度配列２３Ｇに設定された好感度Ｋの平均値を求める。例えば、図２７の例では、体験項目Ｔ１に関して、好感度１−１乃至好感度１−４の平均値が求められ、体験項目Ｔ２に関して、好感度２−１乃至好感度２−８の平均値が求められる。さらにCPU２１は、算出した好感度Ｋの平均値の大きい順番に、体験項目Ｔを並べ替え、図２８に示すような合成体験好感リスト２３Ｈを作成する。なお、このステップＳ３０７の処理の詳細は、図２９のフローチャートを参照して後述する。
【０１２９】
このように、合成体験好感リスト２３Ｈが作成されると処理は終了される。
【０１３０】
次に、図２９のフローチャートを参照して、図２５のステップＳ３０７における合成体験好感リスト作成処理の詳細について説明する。ステップＳ５０１においてCPU ２１は、変数Ｎに合成体験好感リスト２３Ｈの長さを設定し、その代入済み部分のリスト長を表わす変数Ｉに１を設定する。次に、ステップＳ５０２においてCPU ２１は、 av_min に最低好感度限界値を設定する。なお、最低好感度限界値が特に存在しない場合には、 av_min に０を設定してもよい。ステップＳ５０３においてCPU ２１は、体験好感累積バッファ２３Ｅ（図２７）の先頭のポインタを変数Ｋに設定する。
【０１３１】
さらに、ステップＳ５０４においてCPU ２１は、変数ＫがNULLであるか否かを判定する。変数Ｋの値がNULLではない場合、ステップＳ５０５に進み、CPU ２１は、変数Ｋで表わされる体験項目に付随する好感度配列２３Ｇの好感度の平均値を計算する。この処理の詳細は、図３０のフローチャートに示されている。
【０１３２】
すなわち、ステップＳ５３１においてCPU ２１は、合計配列長と好感度累積値に、それぞれ０を設定する。ステップＳ５３２において、CPU ２１は、配列走査ポインタに、配列の先頭の要素を設定する。例えば、図２７の例において、好感度１−１に配列走査ポインタの値が設定される。
【０１３３】
ステップＳ５３３において、CPU ２１は、合計配列長に１を加算し、好感度累積値に、配列走査ポインタが指す配列要素の好感度の値を加算する。今の場合、好感度１−１の好感度の値が好感度累積値に加算される。次に、ステップＳ５３４において、CPU ２１は、配列走査ポインタに配列の次の要素を設定する。今の場合、好感度１−２が配列走査ポインタに設定される。
【０１３４】
次に、ステップＳ５３５において、CPU ２１は、配列に次の要素が存在するか否かを判定する。今の場合、次の配列が存在するため、ステップＳ５３３に戻り、CPU ２１は、再び合計配列長に１を加算し、好感度累積値に配列走査ポインタが指す配列要素の好感度（今の場合、好感度１−２の好感度）の値を累積加算する。ステップＳ５３４に進み、CPU ２１は、配列走査ポインタに配列の次の要素を設定する。今の場合、好感度１−３の配列が配列走査ポインタに設定される。ステップＳ５３５において、CPU ２１は、配列に次の要素が存在するか否かを判定し、存在する場合にはステップＳ５３３に戻り、それ以降の処理を繰り返し実行する。
【０１３５】
以上のようにして、ステップＳ５３５において、配列に次の要素が存在しないと判定されるまで同様の処理が繰り返される。
【０１３６】
今の場合、好感度１−１乃至好感度１−４の好感度の値が累積加算されると、ステップＳ５３５においてNOの判定が行われ、ステップＳ５３６に進み、CPU ２１は、次式から平均好感度を計算する。
【０１３７】
平均好感度＝好感度累積値／合計配列長 （９）
すなわち、今の例の場合、好感度１−１乃至好感度１−４の好感度の累積加算値を、合計配列長である４で割り算して得られた値が平均好感度とされる。そして、この平均好感度の値が変数ave に設定される。
【０１３８】
以上のようにして、平均好感度ave が計算されると、図２９のステップＳ５０６に進み、CPU ２１は、ステップＳ５０５で計算した平均好感度ave の値が、ステップＳ５０２で設定した最低好感度限界値 av_min より大きいか否かを判定する。平均好感度ave が最低好感度限界値 av_min と等しいか、それより小さい場合には、その体験項目は処理対象とする意味がないのでステップＳ５０７に進み、CPU ２１は、変数Ｋに次のポインタの値を設定する。そして、ステップＳ５０４に進み、それ以降の処理が繰り返し実行される。
【０１３９】
ステップＳ５０６において、ステップＳ５０５で計算した平均好感度ave の値が、ステップＳ５０２で設定された最低好感度限界値 av_min より大きいと判定された場合、その体験項目を合成体験好感リスト２３Ｈに登録する処理が以下に実行される。すなわち、ステップＳ５０８において、CPU ２１は、合成体験好感リスト２３Ｈの先頭以降であって、かつＩ番目以前の体験好感要素データＥの好感度を、平均好感度ave と順に比較し（合成体験好感リスト２３Ｈに既に登録されている平均好感度を、いま計算された平均好感度ave と順に比較し）、最初に平均好感度ave 未満となる体験好感要素データＥの位置を変数Ｊに設定する。ステップＳ５０９において、CPU ２１は、Ｊ番目の直前の位置に、今回のＫ番目の体験項目と、その平均好感度ave の値を対にして（体験好感要素データとして）、合成体験好感リスト２３Ｈに挿入する。例えば、合成体験好感リスト２３Ｈの末尾のＩ（ＩがＮの時はＮ−１）番目から先頭のＪ番目までの各体験好感要素データを１づつ後方に（好感度の値が小さい方に）ずらした後、今回のＫ番目の体験項目と平均好感度ave の値が対にして（体験好感要素データとして）挿入される。
【０１４０】
次に、ステップＳ５１０に進み、合成体験好感リスト２３Ｈの代入済み部分のリスト長Ｉが、全体のリスト長Ｎより小さいか否かが判定され、YES の場合、ステップＳ５１１に進み、CPU ２１は、変数Ｉを１だけインクリメントした後、ステップＳ５０７に進み、変数Ｋに次のポインタの値を設定し、ステップＳ５０４に戻り、それ以降の処理を実行する。すなわち今の場合、図２７において、上から２番目の体験項目に関して、上述した体験項目１の場合と同様の処理が実行される。
【０１４１】
以上のようにして、ステップＳ５１０において、合成体験好感リスト２３Ｈの代入済み部分のリスト長Ｉの値が、全体のリスト長Ｎの値と等しいと判定された場合、ステップＳ５１２に進み、CPU ２１は、 av_min に、Ｎ番目の体験好感要素データの好感度を設定する。すなわち、これにより、Ｎ番目の体験好感要素データの好感度と等しいか、それより小さい好感度の体験好感要素データが、合成体験好感リストに登録されないようにする。そして、ステップＳ５０７に戻り、変数Ｋに次のポインタの値が設定され、さらに、ステップＳ５０４以降の処理が繰り返し実行される。処理はステップＳ５０４において、変数ＫがNULLであると判定されるまで繰り返し実行され、変数ＫがNULLと等しいと判定された場合、処理は図２５のステップＳ３０７の合成体験好感リスト作成が完了する（図１６のステップＳ７５に進む）。
【０１４２】
以上のようにして、合成体験好感リスト２３Ｈ（図２７）が生成される。
【０１４３】
なお、ステップＳ３０３の処理において、合成対象会員の個人体験好感レコードＲに記憶されている体験好感要素データＥを、種々の条件に基づいて選別して合成対象会員体験好感リスト２３Ｆに設定することができる。例えば、１ヶ月以内に投稿された体験好感要素データＥのみが、合成対象会員体験好感リスト２３Ｆに設定されるようにすることができ、このようにすると、その条件を満たす体験好感要素データＥに基づいて、合成体験好感リスト２３Ｈが作成される。この場合、この条件を満たす体験好感要素データＥを保持していない会員は、合成対象会員にはならない。また、体験好感要素データＥを選別する条件として、日時の他、体験場所など様々な条件を設定することができ、さらに複数の条件を複合して設定することができる。
【０１４４】
次に、図１６のステップＳ７５における推薦体験（CD）リスト作成処理の詳細を説明する。ステップＳ７４で作成された合成体験好感リスト２３Ｈに基づいて、体験好感情報提供サーバ１４のCPU２１は、例えば図１２乃至図１４に示すような推薦体験（CD）リスト２３Ｉを作成し、通信部２６を制御し、端末１１に送信させる。
【０１４５】
推薦体験リスト２３Ｉの内容は、合成体験好感リスト２３Ｈの合成好感度（図２８の平均好感度）が、対応する体験項目（CDタイトル）の推薦度としてそのまま用いられたものとされたり（この場合、推薦体験リスト２３Ｉは、図２８に示す合成体験好感リスト２３Ｈと同様のリスト（例えば、図１２に示すようなリスト）となる）、あるいは、推薦度の値を含めず、単にCDタイトルだけのリストとされる（この場合、推薦体験リスト２３Ｉは、図２８に示す合成体験好感リスト２３Ｈの平均好感度を削除したリスト（例えば、図１４に示すようなリスト）となる）。
【０１４６】
さらに、推薦体験リスト２３Ｉに含まれるCDタイトルの数をｎ個に制限したり（この場合、推薦体験リスト２３Ｉは、図２８に示す合成体験好感リスト２３Ｈのうちの、上からｎ個（ｎ＜Ｌ）までのリスト（例えば、図１４に示すようなリスト）となる）、最低推薦度以上（平均好感度の値が予め定めた閾値以上）のCDタイトルを盛り込むようにする（この場合、推薦体験リスト２３Ｉは、図２８に示す合成体験好感リスト２３Ｈのうちの、上からｍ個（ｍは、合成体験好感リスト２３Ｈに含まれるCDタイトルの数Ｌ以下の、平均好感度の値が閾値以上のCDタイトルの数を表す）までのリスト（例えば、図１３に示すようなリスト）となる）こともできる。
【０１４７】
また、推薦体験リスト２３Ｉの体験項目の文字を、推薦度に対応して強調することもできる。さらに、推薦体験リスト２３Ｉが人間に読まれない場合、推薦体験リスト２３Ｉの情報を何らかのバイナリ形式で出力することもできる。
【０１４８】
また、図１６のステップＳ７３までの処理の結果として得られる同好者リスト（図２３または図２４の同好者バッファ２３Ｂ）をもとに、同好者紹介サービスを提供することもできる。例えば同好者紹介サービスの提供する情報にメールアドレスなどの同好者紹介情報を盛り込むこともできる。しかしながら、この場合、プライバシー保護のために、会員は、プライバシー情報の提供に制限を加えることができる。例えば、予め申請した特定の体験において、本人と同好者と見なされた会員からの要求があった場合にのみ、プライバシー情報をその同好者に提供することを許可するようにすることなどができる。実際には、体験好感情報提供サーバ１４において、例えば、プライバシー情報提供制限条件が、会員のメールアドレスに対応して記憶され、同好者紹介リストが作成される際、同好者のメールアドレスに対応するプライバシー情報提供制限条件が確認され、その条件が満たされている場合のみ、そのプライバシー情報が同好者紹介リストに盛り込まれる。
【０１４９】
さらに、上述したプライバシー情報提供制限条件に、例えば、予め申請した特定の体験において同好者であること、かつ、日本語を書くことができることなど、複数の条件を複合して設定することができる。なお、このプライバシー情報提供制限条件は、体験好感情報提供サーバ１４の管理者がいくつかの条件を提供し、会員がそれを選択したり、また、会員自身が、ユーザインタフェースを利用して自由に設定することもできる。
【０１５０】
さて次に、連想体験紹介について述べる。これは図１６のステップＳ７３の同好者抽出処理時参照する体験好感リスト２３Ａ，２３Ｄの作成に使用する分野と、図１６のステップＳ７４の合成体験リストの作成に使用する分野とが異なる点が特徴となる。例えば、音楽記事の推薦体験リストの作成を要求した会員が、音楽記事体験に対する体験好感要素データＥを多く有していない場合、そのままでは、十分な推薦体験（音楽記事）リストが作成されない可能性がある。しかしながら、推薦体験リストの作成を要求した会員が、例えば、CD体験に対する体験好感要素データＥを多く有している場合、はじめに、CD体験に対する体験好感要素データＥに基づいて同好者を抽出し、さらに、抽出された同好者の音楽記事体験に対する体験好感要素データＥに基づいて、合成体験好感リストを作成するようにすれば（すなわち、連想体験に基づいて合成体験好感リストを作成するようにすれば）、その会員は、充実した推薦体験（音楽記事）リストを入手することができる。
【０１５１】
図３１は、このような、連想体験に基づく推薦を行う場合のGUI の例を表わしている。同図に示すように、この例においては、入力部１３１にユーザが既に多くの体験投稿をしている分野を入力するようになされている。また、入力部１３２乃至１３５には、そのユーザが推薦を受けたい情報内容を入力するようになされている。この入力欄１３２乃至１３５に入力する情報は、図１１における入力欄１１１乃至１１４において入力される情報と同様の情報であるので、ここでは説明を省略する。
【０１５２】
この図３１に示すようなGUI に基づいてユーザが入力を行い、連想体験の情報の提供を要求する（依頼ボタン１３６を押す）と、端末１１から体験好感情報提供サーバ１４に対して、図３２に示すような連想依頼データが送出される。この連想依頼データは、図３１のGUI に示す入力欄１３１乃至１３５の入力に対応するデータである。同好分野は、図３１の入力部１３１に入力されたデータ（この例の場合「CD」）に対応しており、推薦希望は入力部１３２に入力されたデータ（この例の場合「書籍」）に対応している。最低好感度限界値av_min は入力部１３３に入力されたデータに対応しており、この例の場合、「是非推薦のみ」が指定されているので、最低好感度限界値av_min は0.5 とされている。個数制限は入力部１３４からの入力に対応しており、好感度は入力部１３５からの入力に対応している。
【０１５３】
体験好感情報提供サーバ１４のCPU ２１は、端末１１から図３２に示すような連想依頼データが伝送されてきた時、図３３のフローチャートに示す処理を実行する。最初にステップＳ６０１において、CPU ２１は連想体験の情報の提供を要求してきた端末１１の会員を注目会員として指定する。ステップＳ６０２において、CPU ２１は、ステップＳ６０１で指定された注目会員の同好分野フィールドに指定されているデータベースにアクセスする。例えば、図３２の例の場合、同好分野としてCDが指定されているので、CPU ２１はハードディスク２７に記憶されているデータベースのうち、図３４に示すように、CDに関する会員個人体験好感データベース１５１にアクセスする。
【０１５４】
ステップＳ６０３において、CPU ２１は、ステップＳ６０２でアクセスしたデータベースを用い、同好者抽出処理を実行する。この同好者抽出処理は、上述した図１７を参照して説明した場合と同様の処理である。すなわち、これにより、注目会員とCDタイトルに関して類似した嗜好を有する同好者が抽出される。
【０１５５】
次に、ステップＳ６０４において、CPU ２１は、注目会員の推薦希望フィールドに指定されているデータベースにアクセスする。今の場合、図３２に示すように、推薦希望フィールドには、「書籍」が指定されているので、CPU ２１は、ハードディスク２７に記憶されているデータベースのうち、図３４に示すように、書籍に関する会員個人体験好感データベース１５２にアクセスする。そして、ステップＳ６０５において、CPU ２１は、合成体験好感リスト生成処理を実行する。この合成体験好感リスト生成処理は、体験項目がCDタイトルではなく書籍タイトルである点を除けば、図２５のフローチャートを参照して説明した場合と同様の処理である。すなわち、この処理により、書籍タイトルに関する合成体験好感リストが作成される。さらに、ステップＳ６０６において、CPU ２１は合成体験好感リストから推薦体験リストを生成し、端末１１に出力する。
【０１５６】
このようにして、この例においては、注目会員が嗜好するCDと同一のCDを嗜好する同好者が抽出され、その同好者が嗜好する書籍が検索され、その書籍が推薦体験リストとして注目会員に提供される。
【０１５７】
このようにして、第１の分野（同好者を探す分野）に関して類似した嗜好を有する同好者たちの第２の分野（推薦依頼分野）に関する体験好感要素データＥに基づいて、推薦体験リスト２３Ｉが作成され、それを要求した会員に提供される。
【０１５８】
このように、会員から投稿される体験好感要素データＥにより、個人体験好感レコードＲが更新されるようにしたので、会員のあらゆる分野の嗜好情報を収集できるとともに、容易にサービス（推薦体験リストの提供サービスなど）の範囲を拡大することができる。
【０１５９】
以上においては、本発明をCD体験に関するサービスを提供する場合を例として説明したが、本発明は、これに限らず、各種のサービスを提供する場合に適用することができる。
【０１６０】
なお、本明細書において、システムの用語は、複数の装置、手段などより構成される全体的な装置を意味するものとする。
【０１６１】
【発明の効果】
本発明によれば、興味や嗜好の傾向に対応したサービス情報を提供することができる。
【図面の簡単な説明】
【図１】本発明を適用した体験好感情報提供サーバのネットワークの構成例を示す図である。
【図２】図１の体験好感情報提供サーバ１４の内部の構成例を示すブロック図である。
【図３】図１の端末１１の内部の構成例を示すブロック図である。
【図４】個人体験好感レコードＲの例を示す図である。
【図５】図１の体験好感情報提供サーバ１４の投稿受付処理を説明するフローチャートである。
【図６】図１の端末１１のGUI の表示例を示す図である。
【図７】図１の端末１１のGUI の表示例を示す図である。
【図８】図１の端末１１のGUI の表示例を示す図である。
【図９】図１の端末１１のメッセージの表示例を示す図である。
【図１０】図１の体験好感情報提供サーバ１４の処理を説明するフローチャートである。
【図１１】図１の端末１１のGUI の表示例を示す図である。
【図１２】図１の端末１１の推薦体験情報の表示例を示す図である。
【図１３】図１の端末１１の推薦体験情報の表示例を示す図である。
【図１４】図１の端末１１の推薦体験情報の表示例を示す図である。
【図１５】個人体験好感レコード更新処理を説明するフローチャートである。
【図１６】推薦CDリスト作成処理を説明するフローチャートである。
【図１７】図１６のステップＳ７３の詳細な処理手順を説明するフローチャートである。
【図１８】注目個人体験好感リスト２３Ａの例を示す図である。
【図１９】同好者バッファ２３Ｂの例を示す図である。
【図２０】比較対象会員体験好感リスト２３Ｄの例を示す図である。
【図２１】図１７のステップＳ１０５の詳細な処理手順を説明するフローチャートである。
【図２２】ルックアップテーブルの例を示す図である。
【図２３】同好者バッファ２３Ｂの他の例を示す図である。
【図２４】同好者バッファ２３Ｂのさらに他の例を示す図である。
【図２５】図１６のステップＳ７４の詳細な処理手順を説明するフローチャートである。
【図２６】図２５のステップＳ３０４の詳細な処理手順を説明するフローチャートである。
【図２７】体験好感累積バッファ２３Ｅの例を示す図である。
【図２８】合成体験好感リスト２３Ｈの例を示す図である。
【図２９】図２５のステップＳ３０７の詳細な処理手順を説明するフローチャートである。
【図３０】図２９のステップＳ５０５の詳細な処理手順を説明するフローチャートである。
【図３１】図１の端末１１のGUI の表示例を示す図である。
【図３２】図３１の表示例に対応して図１の端末１１より出力されるデータのフォーマットを説明する図である。
【図３３】図１の体験好感情報提供サーバ１４の動作を説明するフローチャートである。
【図３４】データベースの例を示す図である。
【符号の説明】
１１ 端末， １２ 基地局， １３ 電話回線， １４ 体験好感情報サーバ， ２１ CPU ２２ ROM， ２３ RAM， ２４ 入力部， ２５ 表示部， ２６ 通信部[0001]
[Field of the Invention]
  The present invention relates to an information processing apparatus and method.,recoding media, And information processing systemIn particular, an information processing apparatus and method that can easily collect and store personal preference information and can easily provide service information corresponding to user preference trends,recoding media, And information processing systemAbout.
[0002]
[Prior art]
For example, product introduction information such as direct mail is generally created and distributed in accordance with individual preference trends. Therefore, conventionally, personal information has been collected through questionnaires and the like, and personal preference trends have been detected based on the analysis results.
[0003]
[Problems to be solved by the invention]
However, in the conventional method described above, it is difficult to create a question for obtaining accurate preference information, or the questionnaire results differ depending on the questionnaire grader, and it is difficult to grasp an accurate preference tendency. There was a problem. In addition, there is a problem that it takes a lot of time to create, score, and analyze a questionnaire.
[0004]
Furthermore, there was a problem that private information collected by answering a questionnaire leaked to the outside.
[0005]
The present invention has been made in view of such a situation, and easily collects and accumulates personal preference information, and further provides service information corresponding to user preference trends. is there.
[0006]
[Means for Solving the Problems]
  The information processing apparatus according to claim 1, a storage means for storing a personal information list having a plurality of personal information including experience information about information of each user who uses a plurality of terminal devices, and one personal information to be noted By comparing the experience information contained in the personal information with the experience information contained in each personal information included in the personal information list, the degree of relevance to the personal information of interest from among the personal information included in the personal information list is high. Personal information extraction means for extracting personal information, selection means for selecting information for the terminal device corresponding to the personal information of interest based on the personal information extracted by the personal information extraction means, and information selected by the selection means Presentation control means for controlling the presentation ofThe storage means is provided for each field having a plurality of items, and has a plurality of storage parts for storing a personal information list for each field, and the personal information extraction means includes information among the plurality of storage parts. Personal information of interest based on a plurality of personal information stored in one storage unit in which personal information of the user of the terminal device is stored in the first field input in the terminal device requesting selection of The personal information of the user of the terminal device corresponding to the personal information extracted by the personal information extraction means is used by the personal information extraction means among a plurality of storage units. Different from one storage unit, specified from a plurality of personal information stored in another storage unit related to the second field input in the terminal device that requests selection of information, and based on the specified personal information To select the broadcast.
[0007]
The presentation control means can control the presentation of the information selected by the selection means so as to display in different formats according to the conditions specified by the user of the terminal device that requests the presentation of information.
[0008]
  Presentation controlMeans selected by the selection means via the networkControl presentation of informationTo be able to.
[0009]
  The experience information isinformationItem information indicating these items can be included.
[0010]
  The selection means includes item information included in personal information of interest;IdentifiedInformation can be selected based on the shareability with item information included in personal information.
[0011]
  The experience information isinformationIndicated by item information indicating item and item informationinformationFavorability information showing the likability of itemsWhenCan be included.
[0012]
  The selection means includes item information included in personal information of interest;IdentifiedThe matching status with the item information included in the personal information, and the preference information included in the personal information of interest;IdentifiedInformation can be selected based on the likability information in which the corresponding item information matches among the likability information included in the personal information.
[0013]
Personal information receiving means for receiving personal information transmitted from the plurality of terminal devices may be further provided.
[0014]
  The selection means is identifiedCombining multiple personal informationThe, Generate synthetic personal informationShiBased on the combined personal information, information for the terminal device corresponding to the personal information of interest can be selected.
[0016]
  The selection means includesIdentifiedBased on personal informationTheLimit the information to be selected to the predetermined number of informationRuCan be.
[0017]
  Claim11The information processing method described in 1) includes a storage step in which a computer hard disk stores a personal information list having a plurality of personal information including experience information related to information of each user who uses a plurality of terminal devices, and one individual to be noted By comparing the experience information included in the information with the experience information included in each personal information included in the personal information list, the personal information of interest and the degree of relevance of each personal information included in the personal information list are compared. A personal information extraction step in which the computer CPU extracts high personal information, and a selection step in which the computer CPU selects information for the terminal device corresponding to the personal information of interest based on the personal information extracted in the personal information extraction step And the selection stepsoA presentation control step in which the CPU of the computer controls the presentation of the selected information.The storage step is provided for each field having a plurality of items, and the personal information list for each field is stored in a plurality of storage units provided in the hard disk of the computer. Based on a plurality of pieces of personal information stored in one storage unit in which personal information of the user of the terminal device is stored in the storage unit, the personal information of the user of the terminal device being input in the terminal device that requests selection of information Personal information having a high degree of relevance to the personal information to be noticed, and in the selection step, the personal information of the user of the terminal device corresponding to the personal information extracted in the personal information extraction step Different from the one storage unit used in the information extraction step, it is stored in another storage unit related to the second field input in the terminal device that requests selection of information. It was identified from a plurality of personal information that selects the information based on the identified personal information.
[0018]
  Claim12The recording medium described in 1) stores a personal information list having a plurality of personal information including experience information about information of each user who uses a plurality of terminal devices in a hard disk of the computer, and one piece of personal information to be noted On the other hand, by comparing the experience information included in one personal information of interest with the experience information included in each personal information included in the personal information list, the personal information included in the personal information list A personal information extraction step for causing the CPU of the computer to extract personal information highly relevant to the personal information of interest, and information on the terminal device corresponding to the personal information of interest based on the personal information extracted in the personal information extraction step Selection step to select the computer CPU and selection stepsoA presentation control step for causing the computer CPU to control the presentation of the selected information.The storage step is provided for each field having a plurality of items, and the personal information list for each field is stored in a plurality of storage units provided in the hard disk of the computer. Based on a plurality of pieces of personal information stored in one storage unit in which personal information of the user of the terminal device is stored in the storage unit, the personal information of the user of the terminal device being input in the terminal device that requests selection of information Personal information having a high degree of relevance to the personal information to be noticed, and in the selection step, the personal information of the user of the terminal device corresponding to the personal information extracted in the personal information extraction step Different from the one storage unit used in the information extraction step, it is stored in another storage unit related to the second field input in the terminal device that requests selection of information. It was identified from a plurality of personal information that selects the information based on the identified personal informationRecords programs that cause computers to execute processes.
[0019]
  Claim13A plurality of terminal devices of the information processing system described in the above, input means for inputting experience information related to service information, request information for requesting service information from the information processing device, and experience information and request information via the network. A first communication unit that transmits to the information processing device and receives service information from the information processing device; and a display unit that displays the service information received by the first communication unit. The second communication means for receiving the experience information and the request information from the terminal device and transmitting the service information to the plurality of terminal devices; and the first communication means for the plurality of terminal devices received by the second communication means. Storage means for storing a personal information list having a plurality of pieces of personal information including experience information relating to service information transmitted from the terminal, and a terminal device that has transmitted request information Attention personal information extraction means for extracting the experience information included in the attention personal information that is human information from the storage means, the experience information included in the attention personal information, and each individual included in the personal information list stored in the storage means Related personal information extraction means for extracting personal information highly relevant to the personal information of interest from each personal information included in the personal information list by comparing with the experience information included in the information, and related personal information extraction Selection means for selecting service information for the terminal device corresponding to the personal information of interest based on the personal information extracted by the means, and a presentation control means for controlling the presentation of the service information selected by the selection means.The storage means is provided for each field having a plurality of items, and has a plurality of storage parts for storing a personal information list for each field, and the related personal information extraction means includes a plurality of storage parts, Attention personal information based on a plurality of personal information stored in one storage unit in which personal information of the user of the terminal device is stored in the first field input in the terminal device that requests selection of information The personal information of the terminal device user corresponding to the personal information extracted by the related personal information extracting means is extracted by the selecting means from the plurality of storage units. Different from the one storage unit used, it is specified from a plurality of personal information stored in another storage unit related to the second field input in the terminal device that requests the selection of information, and the specified personal information Zui to select the information.
[0020]
  The information processing apparatus according to claim 1,11Information processing method according to claim 1, claim12And the recording medium according to claim 113In the information processing system described in the above, the personal information included in the personal information list is compared by comparing the experience information included in one personal information of interest with the experience information included in the personal information included in the personal information list. Personal information having a high degree of association with the personal information of interest is extracted. Further, information for the terminal device is selected based on the extracted personal information, and the presentation of the selected information is controlled.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a network connection example of a server for providing experience-preferred information to which the present invention is applied. The terminal 11 conforms to the PIAFS (Personal Handy-phone System Internet Access Forum Standard) standard, uses the user interface to reliably notify the user's intention to the experience-preferred information providing server 14, and also provides the experience-preferred information providing server. 14 is a communication terminal provided with necessary and sufficient functions for reliably transmitting a notification from 14, such as a bitmap display device and a touch panel (both not shown). The terminal 11 communicates with the base station 12 wirelessly via a PHS (Personal Handy-phone System) as a built-in interface, and experiences experience information providing server 14 via a telephone line 13 to which the base station 12 is connected. And can receive services from the experience favorable information providing server 14.
[0022]
The experience preference information providing server 14 is connected to the telephone line 13.
[0023]
The above-described user interface of the terminal 11 can also be constructed in the terminal 11, but when the terminal 11 has a Web viewer function, it is transmitted as HTML data from the experience preference information providing server 14. It is also possible to use.
[0024]
FIG. 2 shows an example of the internal configuration of the experience favorable information providing server 14 of FIG. The CPU 21 performs various processes according to programs stored in the ROM 22 or the hard disk 27.
[0025]
Various programs and data are stored in the ROM 22. The RAM 23 stores programs and data in a preset area corresponding to the processing executed by the CPU 21.
[0026]
The input unit 24 including a keyboard and a mouse is operated by an administrator of the experience preference information providing server 14 when a command is input to the CPU 21. The display unit 25 is configured by a liquid crystal display, for example, and displays various information.
[0027]
The communication unit 26 receives various requests and data from the terminal 11 via the telephone line 13 and outputs data supplied from the CPU 21 to the telephone line 13.
[0028]
In addition to storing the server program, the hard disk 27 stores a personal experience sensation record R for each member (user of the terminal 11) who subscribes to the experience sensation information management system managed by the experience sensation information management company.
[0029]
FIG. 3 shows an internal configuration example of the terminal 11 of FIG. The CPU 51 performs various processes in accordance with programs stored in the ROM 52 or the hard disk 57.
[0030]
The ROM 52 stores various programs and data. The RAM 53 stores programs and data in a preset area corresponding to processing executed by the CPU 51.
[0031]
The input unit 54 including a keyboard and a mouse is operated by the user of the terminal 11 when inputting a command to the CPU 51. The display unit 55 is configured by a liquid crystal display, for example, and displays various types of information.
[0032]
The communication unit 56 is configured by a PIAFS interface, receives various information from the experience-preferred information providing server 14 via the network 13 and the base station 12, and outputs data supplied from the CPU 51 to the base station 12. Has been made.
[0033]
The hard disk 57 stores a program and information provided from the experience preference information providing server 14.
[0034]
FIG. 4 shows an example of information stored in the personal experience feeling record R of the user A in the database formed on the hard disk 27 of the experience feeling information providing server 14. The personal experience feeling record R includes a plurality of experience feeling element data E and experience feeling management data C. Experience sensation element data E is composed of experience item T (title of CD (Compact Disc)) that the member actually experienced (heard) and likability K corresponding to the experience item. In the case of this example, the likability K with respect to the CD is divided into 11 steps from the value 0.0 representing the worst likability to the value 1.0 representing the best likability, and digitized. For example, the preference K of the CD title 201 as the experience item T is 0.9.
[0035]
In the case of this example, seven experience feeling element data E1 to E7 are stored in the personal experience feeling record R, but the experience feeling element data Ei (in this example, i = 8, 9, 10,... When the member is posted to the experience sensation information providing server 14, the personal experience sensation record R is updated, and new experience sensation element data Ei is stored. As the experience sensation element data E, as shown in FIG. 4, the date and time (posting date and time) stored in the personal experience sensation record R and other attributes can also be stored. For example, in the example of FIG. 4, the posting date of the CD title 201 is registered as 10:05 on October 10, 1998.
[0036]
Further, for the reason of managing the storage capacity of the hard disk 27, the maximum number of experience sensation element data E that can be stored in the personal experience sensation record R (hereinafter referred to as the maximum number Neexp of experience sensation element data E) Are predetermined and limited.
[0037]
The experience sensation management data C includes a member ID of a member who is a provider of the experience sensation element data E and an experience item number Ni indicating the number of experience sensation element data E. In this example, the experience sensation management data C stores the ID of the user A as a member ID and 7 as the number of experience items Ni.
[0038]
Next, the process when registering as a member in the experience-preferred information management system managed by the experience-preferred information management company will be described.
[0039]
When receiving the new registration request from the terminal 11 (user), the CPU 21 of the experience sensation information providing server 14 provides the terminal 11 (user) with information such as a contract and a system operating principle, for example. When the CPU 21 of the experience sensation information providing server 14 receives, for example, a digitally signed contract or personal identification information such as a password from the terminal 11, the CPU 21 executes a necessary credit check (credit check) for the user and confirms the confirmation. The received information is stored in the temporary RAM 23 until it can be taken. When the credit confirmation result is OK, the CPU 21 of the experience favorable information providing server 14 gives a password to the terminal 11 (user) and officially stores the information stored in the RAM 23 in the database of the hard disk 27. Register and finish new user registration.
[0040]
As described above, the user (terminal 11) registered in the experience preference information providing server 14 can post various experience preference information to the experience preference information providing server 14. The processing in this case will be described with reference to the flowchart of FIG.
[0041]
The posting user operates the input unit 54 of the terminal 11 to activate a browser for accessing the experience preference information providing server 14. At this time, the CPU 51 displays a GUI (Graphical User Interface) for access as shown in FIG. The user inputs a URL (Uniform Resource Locator) for accessing the experience preference information providing server 14 in the connection destination input field 71. Further, when the start of connection is instructed, the CPU 51 controls the communication unit 56 to execute access to the URL (experience preference information providing server 14) input in the input field 71.
[0042]
Upon receiving access from the terminal 11, the CPU 21 of the experience sensation information providing server 14 activates the incoming call processing program, controls the communication unit 26 in step S 11, and displays HTML (Hipertext Markup Language) data on the login (Login) screen. Is sent to the terminal 11. When the login screen transmission process is completed, the CPU 21 once terminates the operation of the incoming call processing program.
[0043]
When the CPU 51 of the terminal 11 receives the HTML data of the login screen via the communication unit 56, the CPU 51 processes this, and causes the display unit 55 to display a posting screen GUI as shown in FIG. The user operates the input unit 54 to input his / her name in the input field 81 and also inputs the password assigned to himself by the registration process described above in the input field 82. When posting, the user further operates the mouse of the input unit 54 to turn on the posting button 83. In order to receive provision of recommendation information, the user operates the recommendation button 84. In order to return the GUI to the original screen, for example, the service overview comment screen, the return button 85 is operated.
[0044]
For example, now, assuming that the user has entered the name and password in the input fields 81 and 82 and then operated the posting button 83, the CPU 51 of the terminal 11 attaches the name and password via the communication unit 56. And output a request for post acceptance processing.
[0045]
When the CPU 21 of the experience sensation information providing server 14 receives this request via the communication unit 26, it activates the incoming call processing program again, and executes a process of confirming and authenticating the user in step S21. Specifically, in step S22, the CPU 21 determines whether or not the input user name and password are registered in the database of the hard disk 27 (whether or not the user confirmation and authentication result is OK). If it is determined and not registered (if the user confirmation and authentication result is not OK), the process returns to step S11, and the process of sending the login screen to the user is executed again.
[0046]
On the other hand, when the received name and password are registered in the database of the hard disk 27 (when the user confirmation and the authentication result are OK), the CPU 21 proceeds to step S23 to post the experience. The HTML data of the business form is transmitted to the terminal 11, and the process is terminated.
[0047]
When the CPU 51 of the terminal 11 receives the HTML data of the experience posting form via the communication unit 56, the CPU 51 processes this, and causes the display unit 55 to display, for example, a GUI as shown in FIG. In this GUI, a desired field is selected from fields such as CD, video, and book in the field selection field 91. The user selects a field of experience to post from these. For example, as shown in FIG. 8, when a field for posting a CD is selected in the input field 91, the user inputs a CD title in the input field 92. Furthermore, the user inputs the favorable feeling felt by the user in the input column 93 for the title CD input in the input column 92. In this example, one good sensitivity can be selected from good, medium, and bad. The user operates the posting button 94 when the input in the input fields 91 to 93 is completed. At this time, the CPU 51 controls the communication unit 56, attaches the input experience posting data and the member ID, and outputs a request for registration processing of the experience posting data to the experience preference information providing server 14.
[0048]
When the CPU 21 of the experience sensation information providing server 14 receives this request via the communication unit 26, the CPU 21 activates the incoming call processing program, and executes the parsing process of the user post data received from the terminal 11 in step S31. To do. That is, the CPU 21 determines whether or not predetermined items are correctly input in the three input fields 91 to 93 shown in FIG. 8, and if they are not correctly input, the CPU 21 returns to step S23, and the experience posting form The process of sending the HTML data to the terminal 11 again is executed.
[0049]
If the CPU 21 determines in step S32 that correct input has been performed, the CPU 21 proceeds to step S33, controls the communication unit 26, and sends out the HTML data of the posting acceptance message.
[0050]
When the CPU 51 of the terminal 11 receives the HTML data of the posting acceptance message via the communication unit 56, it processes this and causes the display unit 55 to display, for example, a GUI as shown in FIG. In this example, the field (CD in this example) designated by the user in the input field 91 shown in FIG. The display column 102 displays the CD title input by the user in the input column 92 of FIG. Furthermore, the display column 103 displays the likability input by the user in the input column 93 of FIG. In this way, the user can confirm that the experience preference information input by himself / herself has been correctly posted on the experience preference information providing server 14.
[0051]
On the other hand, in step S34, the CPU 21 of the experience sensation information providing server 14 has the number Ni of experience items of the user existing in the member individual experience sensation basic knowledge database (DB) of the hard disk 27 within the maximum stored number Nemax. It is determined whether or not there is. If the number of experience items Ni has not yet reached the maximum number of stored items Nemax, the process proceeds to step S36, and the CPU 21 executes a process of additionally inserting the user's post data posted this time from the terminal 11 into the database. To do. In step S34, when it is determined that the user's experience sensation element data E is already registered in the database by the maximum storage number Nemax, the process proceeds to step S35, and the CPU 21 refers to the posting date to Delete one of the oldest experience feeling element data E of the user in the database. In step S36, the CPU 21 executes a process of additionally inserting the current posting data in place of the deleted experiential element data E.
[0052]
As described above, the user can post and register his / her own personal experience preference information from the terminal 11 to the experience preference information providing server 14 and register it.
[0053]
As described above, a user who has posted his / her own experience preference information (your own preference) can receive information (service) that matches his / her preference from the experience preference information providing server 14. . In this case, the user operates the recommendation button 84 instead of the posting button 83 in FIG. At this time, the CPU 21 of the experience preference information providing server 14 executes the process shown in the flowchart of FIG. 10 instead of the process after step S21 of FIG.
[0054]
As shown in the flowchart of FIG. 10, when the confirmation and authentication result is OK, the CPU 21 of the experience preference information providing server 14 does not use the HTML data of the experience posting form but the recommended experience list request form in step S23. Are sent to the terminal 11. As a result, the CPU 51 of the terminal 11 causes the display unit 55 to display a GUI screen for requesting a recommended experience list as shown in FIG. 11, for example.
[0055]
In the display example of FIG. 11, the field of information that the user wants to receive recommendation, such as CD, video, and book, is input to the input field 111 in the recommendation request conditions. In the input field 112, it is input whether the condition of the output range is “all” or “recommended only”. The maximum number of pieces of information such as CDs, videos, books, etc. recommended by the experience-preferred information providing server 14 to the terminal 11 is input to the input unit 113. In the input field 114, any one of “no display”, “numerical value only”, “in words”, or “both” is displayed as the method of displaying the favorable impression when the experience sensation information providing server 14 makes a recommendation. It has been made so that you can select and input. The request button 115 is operated when input is completed. When receiving the request for the recommendation data generation request attached with the password from the terminal 11, the CPU 21 of the experience preference information providing server 14 activates the incoming call processing program, and in step S41, the condition of the recommendation request received from the terminal 11 is satisfied. Parse the syntax. In step S42, the CPU 21 determines whether or not the syntax analysis result is correct. If it is not correct, the CPU 21 returns to step S23 and executes a process of sending out the HTML data of the recommended experience list request form again. If it is determined in step S42 that the syntax analysis result is correct, the process proceeds to step S43, and the CPU 21 executes a process of generating and sending a recommended experience list.
[0056]
12 to 14 show an example of the recommended experience list generated by the CPU 21 of the experience preference information providing server 14 in step S43. FIGS. 12 to 14 are examples in which a CD is designated as the field of information that the user wants to receive recommendations in the input field 111 of FIG.
[0057]
The display example of FIG. 12 represents a display example when the user selects the item “both” in the favorableness display column 114 of FIG. 11. That is, in this case, the recommendation is made by both words and numbers. In other words, if the item “numerical value only” is selected, only the numerical value “1.0” is displayed as the recommendation level corresponding to the CD title “AAAAA”. In addition, when the item “in words” is selected as the preference, only the word “indefinitely” is displayed as the recommendation level. On the other hand, when the item “both” is selected, as shown in FIG. 12, the recommendation level is displayed in both words and numerical values, such as “Needly (1.0)”. In the example of FIG. 12, since the “all” item is selected as the output range of the input field 112 of FIG. 11, all recommended CDs are displayed.
[0058]
FIG. 13 shows a display example of recommendation data when “recommendation only” is selected as the output range of the input field 112 and the item “no display” is selected as the favorable sensitivity in the GUI of FIG. Yes. In this example, since the item “Recommended only” is selected, CDs with a low recommendation level such as “Normal”, “Introduction”, and “Reference” in FIG. 12 are not introduced. Only CDs with a high recommendation level of “Please come” are introduced. In addition, since the “no display” item is selected in the input field 114, only the CD title is displayed, and the recommendation level is not displayed in either numerical values or words.
[0059]
FIG. 14 shows a display example of recommended data when “5” is input as the number limit in the input column 113 and the item “No display” is selected as the favorable rate in the input column 114 in the GUI of FIG. It represents. In this example, five CD titles are selected in order from the one with the highest recommendation value.
[0060]
Of the above processing, the processing procedure of the CPU 21 of the experience sensation information providing server 14 when the personal experience sensation record R is updated will be described in more detail with reference to the flowchart of FIG.
[0061]
As described above, the user A who has already registered as a member in the experience preference information management system and has the personal experience preference record R recorded in the database of the hard disk 27 as shown in FIG. 11, accessing the experience sensation information providing server 14 via the telephone line 13 and transmitting (posting) new experience sensation element data Ei with a member ID, provision of experience sensation information is provided in step S51. The CPU 21 of the server 14 receives new experience sensation element data Ei transmitted from the terminal 11 via the communication unit 26 and temporarily stores it in the RAM 23.
[0062]
Next, in step S52, the CPU 21 determines whether or not the user A who is the transmission source of the experience likability element data Ei is a member of the experience sensation information management system, and the member ID transmitted by the user A is the personal experience. If it is determined by comparing with the member ID (FIG. 4) stored in the likability record R and it is determined that the user A is a member of the experience sensation information management system, the process proceeds to step S53. In step S53, the CPU 21 determines whether the transmitted experience sensation element data Ei is, for example, whether or not the transmitted experience sensation element data Ei includes both data of the experience item T and likability K. It is determined whether or not the data is appropriately formatted so that it can be added to the personal experience sensation record R.
[0063]
If it is determined in step S53 that the experience sensation element data Ei received in step S51 is appropriately formatted data, the CPU 21 stores the personal experience sensation record of the user A stored in the hard disk 27 in step S54. In R, it is determined whether or not the experience feeling element data Ex for the maximum stored number Nemax is stored. If it is determined that the experience feeling element data Ex for the maximum stored number Nemax is stored, the process proceeds to step S55. .
[0064]
In step S55, the CPU 21 detects the experience sensation element data Ek stored most recently (before time) from the experience sensation element data Ex stored in the personal experience sensation record R of the user A. Next, in step S56, the experience feeling element data Ei received in step S51 is stored in the personal experience feeling record R of the user A on the hard disk 27. That is, the experience feeling element data Ei received in step S51 is stored in the personal experience feeling record R in place of the experience feeling element data Ek deleted in step S5.
[0065]
When the CPU 21 determines in step S54 that the number of experience items Ni stored in the personal experience likability record R of the user A has not reached the maximum stored number Nemax, the CPU 21 proceeds to step S56 and received in step S51. The experience sensation element data Ei is added and stored in the personal experience sensation record R of the user A as it is.
[0066]
If it is determined in step S52 that the user A is not a member of the experience preference information management system, or if it is determined in step S53 that the experience preference element data Ei received in step S51 is not properly formatted, In step S57, the CPU 21 controls the communication unit 26 to notify the terminal 11 of the fact. Based on this notification, the user performs the same posting procedure again. Note that the experience sensation element data Ei transmitted from the user A stored in the RAM 23 in step S51 is erased from the RAM 23 at the same time as the processing is completed.
[0067]
In this way, the experience feeling element data Ei posted by the member is additionally recorded, and the individual experience feeling record R of the member is updated.
[0068]
In the above, in order to manage the resources (storage capacity) of the hard disk 27, the maximum stored number Nemax of experience sensation element data E that can be stored in the personal experience sensation record R is set, and the experience sensation element data E is stored. Based on the time, the newly posted experience feeling element data E is replaced with the old experience feeling element data E. However, if the resources of the hard disk 27 can be managed, other methods can be used. For example, news article information becomes meaningless to the user as the years pass. Therefore, when the experience sensation element data E has a news article as an experience item, the longest storage period stored in the personal experience sensation record R is set, and all the experience sensation element data E stored beyond that is deleted. It can also be done.
[0069]
In addition, the member can also post a plurality of experience favorable element data E by operating the terminal 11 based on the corresponding user interface.
[0070]
Next, the processing procedure of the experience preference information providing server 14 when creating the recommended experience list will be described in more detail by taking as an example the case of receiving a recommendation of a CD that matches the user's preference. This will be described with reference to the flowchart of FIG.
[0071]
The user A who has already registered the personal experience sensation record R shown in FIG. 4 in the experience sensation information providing server 14 operates the input unit 54 of the terminal 11 (CD is input in the GUI input field 111 shown in FIG. 11). When the designated item is designated in the other input fields 112 to 114 and the request button 115 is turned on), the CPU 51 of the terminal 11 controls the communication unit 56 to send a recommendation CD list creation request to the telephone. It is transmitted to the experience sensation information providing server 14 via the line 13. The CPU 51 of the terminal 11 includes the member ID of the user A in this request. In step S <b> 71, the CPU 21 of the experience preference information providing server 14 receives the recommended CD list creation request transmitted from the terminal 11 via the communication unit 26.
[0072]
Next, in step S <b> 72, the CPU 21 determines the member ID of the user A who transmitted the recommended CD list creation request as the member of interest, and stores the member ID of the member of interest in the RAM 23.
[0073]
Next, in step S73, the CPU 21 extracts a member having a preference similar to the user A who is the member of interest, that is, a member who likes a CD (music) similar to the user A (details of the process will be described later). ).
[0074]
Next, in step S74, the CPU 21 collects the experience sensation element data E of members who have similar preferences as the user A extracted in step S73, and generates a composite experience sensation list (details of the process will be described later). .
[0075]
Next, in step S <b> 75, the CPU 21 creates a recommended experience (CD) list suitable for the user A from the combined experience sensation list generated in step S <b> 74 and transmits it to the terminal 11. Thereafter, the process is terminated.
[0076]
In this way, once a recommended experience (CD) list suitable for the user is created, it is provided to the requesting user.
[0077]
In the following, details of each process in steps S73 to S75 described above will be described. First, the details of the likes extraction process in step S73 will be described with reference to the flowchart of FIG.
[0078]
In step S101, the CPU 21 of the experience favorable feeling information providing server 14 corresponds to the member ID of the member (member who requested creation of the recommended CD list) determined as the member of interest in step S72 (in this case, the user A). The personal experience feeling record R is read from the hard disk 27, and the experience feeling element data E included in the personal experience feeling record R is written in the attention individual experience feeling list 23 A and stored in the RAM 23. FIG. 18 shows an example of the attention individual experience sensation list 23A created in this way. In the case of this example, in the attention individual experience feeling list 23A, experience feeling element data E of the personal experience feeling record R of the user A shown in FIG. 4 is stored with a management number.
[0079]
Next, in step S102, the CPU 21 determines the member ID of a member (hereinafter referred to as a “favorite”) that has been determined to have a preference similar to the user A and the degree of similarity F (described later) (hereinafter, referred to) by processing described below. The member buffer 23B in the RAM 23 is initialized to store the member ID and the degree of similarity F together. When the affiliation buffer 23B is initialized, as shown in FIG. 19, no data is set in the column in which the member ID of the affiliation is set, and in the column in which the similarity F is set, 0 is set as an initial value. Note that the maximum number of likes data stored in the likes buffer 23B (hereinafter referred to as the maximum number Nfmax of likes data stored) is determined in advance.
[0080]
In step S103, the CPU 21 determines a member (hereinafter referred to as a comparison target member) that examines whether or not the user A has a preference similar to that of the user A, that is, whether or not he / she is a similar user, and the member ID of the member For example, in this case, the member ID of the user B is stored in the comparison target member ID buffer 23C (not shown) in the RAM 23 (in this step S103, the registered member ID And a comparative member may be extracted based on information such as a residence area (address).
[0081]
Next, in step S104, the CPU 21 reads from the hard disk 27 the personal experience sensation record R (FIG. 4) of the comparison target member (user B), whose member ID is set in the comparison target member ID buffer 23C in step S103. The experience sensation element data E is stored in the comparison target member experience sensation list 23D of the RAM 23 with a number stored (set). FIG. 20 shows an example of the comparison target member experience sensation list 23D. The data stored in the comparison target member preference list 23D corresponds to the experience preference element data E of the personal experience preference record R of the user B who is the comparison target member. In this example, the CD heard by the user B CD titles 201 to 204, CD titles 301 to 305, and favorable sensitivity K for each CD.
[0082]
In step S105, the CPU 21 stores the data (experience sensation element data E of user A) stored in the attention member experience sensation list 23A (FIG. 18) of the RAM 23 and the comparison target member experience sensation list 23D (FIG. 20). Based on the recorded data (experience sensation element data E of user B), the similarity F between the member of interest (user A) and the member to be compared (user B) is calculated.
[0083]
Details of the processing for calculating the degree of similarity F in step S105 will be described with reference to the flowchart of FIG.
[0084]
First, in step S201, the CPU 21 of the experience sensation information providing server 14 calculates the experience sharing rate P1 between the target member and the comparison target member based on Expression (1). The experience sharing rate P1 is obtained by dividing the number M of experience items (hereinafter referred to as shared experience item TP) experienced by both the notable member and the comparison target member by the total number of experience items Nm of the notable member. The larger the number of CDs heard by the member of interest and the member to be compared, the larger the value.
[0085]
Experience sharing rate P1 = (number of shared experience items M) / (total number of experience items Nm of the notable member) (1)
As is clear from comparison between FIG. 18 and FIG. 20, in this example, the shared experience items TP for the user A and the user B are the four CD titles 201 to 204, so the number of shared experience items M is 4. Is done. In addition, since the experience item T of the user A who is the member of interest is the seven CD titles 201 to 207, the total number of experience items Nm of the member of interest is 7, and the experience sharing rate P1 is 4/7.
[0086]
Next, in step S202, the CPU 21 determines whether or not the experience sharing rate P1 calculated in step S201 is greater than a preset threshold value. If the experience sharing rate P1 is greater than the threshold value, the process proceeds to step S203. Proceed to
[0087]
In step S <b> 203, the CPU 21 calculates a favorable similarity P <b> 2 between the target member and the comparison target member based on the formula (2). The likability similarity P2 is an index indicating how well the likability KAi evaluated by the member of interest for each shared experience item TP matches the likability KBi evaluated by the comparison target member. The higher the similarity, the higher the value.
[0088]
[Expression 1]

The denominator M of the second term in Expression (2) represents the number of shared experience items. In this example, the number M of shared experience items is 4. The numerator of the second term in the equation (2) is the sum of the sensation differences ei representing the difference in the likability between the target member and the comparison target member for each shared experience item TP. The sensation difference ei is based on the equation (3). Is required.
[0089]
ei = | KAi-KBi | × WA (KAi) × WB (KBi) (3)
KAi and KBi (in the following, when it is not necessary to distinguish between KAi and KBi, the two are collectively referred to as Ki), the preference member and the comparative member of the shared experience item TP respectively. Sensitivity. That is, the difference between the interest member's preference KAi for the shared experience item TP and the comparison member's preference KBi is obtained by | KAi−KBi | which is the first factor on the right side of Expression (3).
[0090]
WA (KAi), the second factor on the right side of equation (3), and WB (KBi), the third factor (In the following, it is not necessary to distinguish between WA (KAi) and WB (KBi). And W (Ki) are collectively described as weighting functions corresponding to the ambiguity of the favorable feeling Ki for the shared experience item TP of the target member and the comparison target member. Calculated based on In the formula (4), 0 ≦ α ≦ 1.
[0091]
W (Ki) = 4α (Ki−0.5) 2+ (1−α) (4)
When the favorable sensitivity Ki is 1.0 or when the favorable sensitivity Ki is 0.0, W (Ki) has a maximum value of 1.0, and when the favorable sensitivity Ki is 0.5, W (Ki) has a minimum value (1-α). Become.
[0092]
When likability Ki is set to 0.5, the shared experience item TP is likable between the best and worst favour, and is evaluated with the most ambiguity that it is neither good nor bad. It is thought that. Therefore, the value of W (Ki) is reduced as the likability Ki is closer to 0.5, that is, the evaluation including more ambiguity is performed, and as a result, the value of the likability difference ei is decreased. Thus, the ambiguity of evaluation is suppressed from being included in the likable similarity P2 finally obtained.
[0093]
In addition, in the weighting function WA (KAi) corresponding to the above-mentioned member of interest and the weighting function WB (KBi) corresponding to the member to be compared, different α values are set, or the evaluation is ambiguous based on different weighting functions. The weight corresponding to can also be calculated. In addition, by setting the α value to 0, it is possible to calculate the favorable similarity P2 that does not consider the weight corresponding to the ambiguity.
[0094]
Note that the likability difference can be calculated by referring to a lookup table as shown in FIG. 22, for example, instead of calculating Expression (3). In this look-up table, the horizontal axis represents the favorable sensitivity KAi, and the vertical axis represents the favorable sensitivity KBi, and a value specified by the two values is obtained as the favorable difference. For example, when the positivity KAi on the horizontal axis is 0.3 and the positivity KBi on the vertical axis is 0.8, the likability difference is 0.4.
[0095]
Returning to FIG. 21, in step S204, the CPU 21 determines whether or not the favorable similarity P2 calculated in step S203 is larger than a preset threshold, and if the calculated favorable similarity P2 is larger than the threshold, In step S205, the preference similarity P2 is set as the similarity F. It should be noted that the preference similarity P2 is adjusted by the experience sharing rate P1, such as the experience sharing rate P1 calculated in step S201 and the value obtained by dividing the preference similarity P2 calculated in step S203 as the similarity F. The degree of similarity F can also be obtained.
[0096]
When it is determined in step S202 that the experience sharing rate P1 calculated in step S201 is smaller than the threshold value, or when the favorable similarity P2 calculated in step 203 is determined to be smaller than the threshold value in step S204 In step S206, the CPU 21 sets the similarity to 0. That is, in this case, the preference of the comparison target member is not similar to the preference of the member of interest.
[0097]
In this way, when the processing for calculating the similarity F shown in FIG. 21 is completed, the process proceeds to step S106 in FIG.
[0098]
In the above description, the experience sharing rate P1 is calculated based on the formula (1), but can also be calculated based on the formula (5).
[0099]
Experience sharing rate P1 = (number of shared experience items M) / (total number of experience items Nmab experienced by at least one member of interest and comparison target member) (5)
In this example, the total number of experience items experienced by the notable member (user A) is 7 (FIG. 18), and the total number of experience items experienced by the comparison target member (user B) is 9 (FIG. 20). Since four of the experience items (CD titles 201 to 204) are shared experience items, the total number of experience items Nmab experienced by at least one of the target member and the comparison target member is 12 (= 7 + 9-4). The experience sharing rate P1 is 4/12. That is, the experience sharing rate P1 obtained by the equation (1) is obtained regardless of the total number of experience items of the comparison target member, but the experience sharing rate P1 obtained by the equation (5) is the total number of experience items of the comparison target member. It depends on.
[0100]
In the above description, the likability similarity P2 is calculated based on the formula (2), but can also be calculated based on the formula (6).
[0101]
[Expression 2]

Of the second term of the formula (6), the value of the numerator represented by the formula (7) indicates the favorable sensitivity KAi evaluated by the member of interest for the shared experience item TP and the favorable sensitivity KBi evaluated by the member to be compared. The closer the value is, the smaller the value. On the other hand, among the second term denominator of the equation (6), the value of the second term represented by the equation (8) is a value in which both the favorable sensitivities KAi and KBi are close, and the favorable sensitivities KAi and KBi are The closer to 1.0 or 0.0, the larger the value. That is, the likability similarity P2 calculated based on the equation (6) is the same when the sensitivities KAi and KBi are close to 0.0 or 1.0 at the same time, that is, the same clearer conclusion (unambiguous conclusion). If it is made, the exchange similarity P2 is weighted so as to have a large value.
[0102]
[Equation 3]

[Expression 4]

Note that the likable similarity P2 can be calculated using covariance. This method calculates the favorable similarity P2 from an eigenvalue corresponding to a positive correlation (eigenvector in the first or third quadrant) and an eigenvalue corresponding to a negative correlation (eigenvector in the second or fourth quadrant). First, a covariance coefficient matrix is obtained from the mean value of both favorable sensitivities and the sum of squares and products of both favorable sensitivities Ki, and by diagonalizing the matrix, two eigenvalues and their corresponding values are obtained. An eigenvector (main direction) is obtained. For example, the ratio of two obtained eigenvalues is taken and a value obtained by subtracting from 1 is set as the favorable similarity P2.
[0103]
Returning to FIG. 17, in step S <b> 106, the CPU 21 stores the maximum number of stored records Nfmax in the likes buffer 23 </ b> B in the RAM 23 (in the example of FIGS. 19, 23, and 24, the maximum stored number Nfmax is equal to L). It is determined whether or not the favorite data is stored, and if the favorite data for the maximum number Nfmax is stored, the process proceeds to step S107, and the similarity F smaller than the similarity F calculated in step S105 is set. It is determined whether or not the included lover data exists in the lover buffer 23B. If such lover data exists, the process proceeds to step S108.
[0104]
In step S108, the CPU 21 deletes the one degree of similarity data having the smallest value of the degree of similarity F from the degree of similarity data that is confirmed in step S107 and includes the degree of similarity F smaller than the degree of similarity F calculated in step S105. To do. Specifically, as shown in FIG. 23, in the likes buffer 23 </ b> B, the likes data having a large value of the similarity F is held so as to have a small number (so as to be positioned at the top in the figure). Has been. Accordingly, the affiliation data having the largest number (number L in FIG. 23) is deleted here. Note that the maximum number of stored items Nfmax in the lover buffer 23B can be changed within the range of the storage capacity of the hard disk 27.
[0105]
Next, in step S109, a process for adding new lover data to the lover buffer 23B is executed. That is, the CPU 21 stores the affiliation data including the affiliation F calculated in step S105 in the affiliation buffer 23B so that the affiliation data having a high affinity F corresponds to a smaller number. Specifically, the CPU 21 compares the degree of similarity F calculated in step S105 with the degree of similarity F of the highest-numbered affiliation data. The process of comparing with degree F is repeated. If the value of the similarity F calculated in step S105 is equal to or smaller than the value of the similarity F stored in the likes buffer 23B, the stored likes are stored. New lover data is stored in the position of the number one larger than the data number.
[0106]
In step S110, the CPU 21 determines whether or not the hard disk 27 has data of a member whose similarity F with the member of interest has not yet been calculated (a member that has not yet been compared). If such member data exists, the process proceeds to step S111, where the member ID of the next member to be compared is selected from the data of members not yet compared, and the member ID buffer to be compared in the RAM 23 is selected. Set to 23C. Thereafter, the process returns to step S104, and the subsequent processes are repeatedly executed.
[0107]
If the CPU 21 determines in step S106 that the maximum number Nfmax of the favorite data is not stored in the favorite buffer 23B (as shown in FIG. 19 or FIG. 24, it is stored in the favorite buffer 23B). When the number of likes data is smaller than L), the process proceeds to step S109, and the likes data including the likes F calculated in step S105 is stored in the likes buffer 23B in the order of the magnitude F. Let
[0108]
If it is determined in step S107 that there is no similarity F having a value smaller than the similarity F calculated in step S105, the CPU 21 skips the processing of steps S108 and S109 (the affiliation data is stored in the affiliation buffer 23B). In step S110, it is determined whether there is a member whose similarity F with the member of interest is not required.
[0109]
If it is determined in step S110 that there is no member for which the similarity F with the member of interest has not yet been calculated, the processing is terminated.
[0110]
In this way, lovers having the same preference as the member of interest (user A) are extracted, and lover data is stored in the lover buffer 23B.
[0111]
In the process of step S104, the experience sensation element data E stored in the comparison member's personal experience sensation record R may be selected and set in the comparison member experience sensation list 23D based on various conditions. it can. For example, only the experience sensation element data E posted within one month can be set in the comparison target member experience sensation list 23D. Based on this, lovers can be extracted. In this case, a member who does not hold the experience sensation element data E satisfying this condition is not a comparison target member. In addition to the date and time, various conditions such as the experience location can be set as conditions for selecting the experience likability element data E, and a plurality of conditions can be combined.
[0112]
Also, as a condition for extracting the likes, in addition to the above-mentioned preference F, whether it belongs to gender, age, address, specific organization, or whether it is possible to contact personally by e-mail etc. Further conditions such as can be added. Actually, in step S103, the CPU 21 determines a comparison target member based on the set likes extraction condition. In this case, for a member that is a candidate for comparison, a composite value is calculated from the logical value obtained for each simple condition proposition according to the theoretical formula corresponding to the compound condition proposition, and this is determined as a judgment value. Or a theoretical value for a compound conditional proposition can be calculated at once and used as a judgment value. Since the comparison target members are selected based on the determination value calculated in this way, members that do not satisfy the likes extraction condition are not extracted as the likes.
[0113]
Further, in the above, the attention member experience sensation list 23A is stored in the RAM 23 because it is repeatedly referred to. For example, the attention member experience sensation list 23A has a configuration shown in FIG. By storing the attention member experience favorable feeling list 23A in the associative memory (contents addressable memory), the processing can be executed at higher speed. For example, out of the data stored in the attention member experience sensation list 23A, the data set in the experience item is set as fixed-length data, and it is set as address data on the associative memory. Then, the likability corresponding thereto is stored in the associative memory as the associative data.
[0114]
In this way, for example, when the experience sharing rate P1 is calculated in step S201, when the attention member experience feeling list 23A is stored in the RAM 23, the comparison member experience feeling list 23D and the attention member experience feeling are compared. Both contents of the list 23A are confirmed. That is, in the case of this example, 63 comparison processes are executed, which is the product of 7 which is the length of the noticeable member experience feeling list 23A and 9 which is the length of the comparison target member experience feeling list 23D. On the other hand, when the attention member experience sensation list 23A is stored in the associative memory, comparison processing for nine times, which is the length of the comparison target member sensation list 23D, is performed.
[0115]
Next, details of the combined experience / favorite list generation process in step S74 of FIG. 16 will be described with reference to the flowchart of FIG.
[0116]
First, in step S301, the CPU 21 initializes an experience sensation accumulation buffer 23E (described later with reference to FIG. 27) in the RAM 23, and then in step S302, the lovers stored in the lover buffer 23D. The value j of the counter for counting the number of data is initialized to 1.
[0117]
In step S303, the CPU 21 sets the member of the member ID included in the affiliation data of the number corresponding to the counter value j as the composition target member, reads the personal experience sensation record R of the member from the hard disk 27, and composes the RAM 23. It memorize | stores in the object member experience favorable feeling list 23F (not shown). When j = 1, among the affiliation data stored in the affiliation buffer 23B, the member of the member ID that is located at the head (the smallest number), that is, the member ID corresponding to the largest degree of similarity F is obtained. Is a member to be synthesized.
[0118]
Next, in step S304, the CPU 21 performs a process of adding data to the experience sensation cumulative buffer 23E based on the experience sensation element data E stored in the compositing target member experience sensation list 23F.
[0119]
The data addition process to the experience sensation accumulation buffer 23E in step S304 will be described with reference to the flowchart of FIG.
[0120]
In step S401, the CPU 21 initializes to 1 the value i of the counter that counts the number of experience items Ni stored in the compositing target member experience preference list 23F.
[0121]
Next, in step S402, the CPU 21 determines whether or not the experience item Ti of the experience likability element data E corresponding to the counter value i exists in the experience item list of the experience likability accumulation buffer 23E, and determines that it exists. If so, the process proceeds to step S403, and the likability K of the experience sensation element data E corresponding to the counter value i is added to the favourability array 23G (FIG. 27) corresponding to the experience item Ti (CD title).
[0122]
In step S402, the CPU 21 finds that the experience item Ti of the experience sensation element data E corresponding to the counter value i (stored in the compositing target member experience sensation list 23F) exists in the experience item list of the experience sensation accumulation buffer 23E. If not, in step S404, the experience item Ti of the experience preference element data E is set at the end of the experience item list in the experience preference accumulation buffer 23E, and in step S405, the preference K corresponding to the experience item Ti. Is set in the favorability array 23G corresponding to the experience item Ti.
[0123]
In step S403 or step S405, when likability K is set in the favourite array 23G, the process proceeds to step S406, and the CPU 21 determines whether or not the counter value i is equal to the number N of experience items in the composition target member experience sensation list 23F. If they are not equal, the process proceeds to step S407, the counter value i is incremented by 1, the process returns to step S402, and the subsequent processing is repeated.
[0124]
If the CPU 21 determines in step S406 that the counter value i is equal to the number N of experience items in the compositing target member experience sensation list 23F, the process ends, and the process proceeds to step 305 in FIG.
[0125]
In step S305, the CPU 21 determines whether or not the value j of the counter that counts the number of likes data stored in the likes buffer 23B is equal to the last number L of the likes buffer 23B. If it is determined that the value j is not equal to the last number L of the lover buffer 23B, the process proceeds to step S306, the counter value j is incremented by 1, the process returns to step S303, and the subsequent processing is repeatedly executed.
[0126]
Thus, as shown in FIG. 27, the experience sensation cumulative buffer 23E is set with the sensibility K for the experience item T and the favorite array 23G corresponding thereto.
[0127]
If the CPU 21 determines in step S305 that the counter value j is equal to the number L of the likes data stored in the likes buffer 23B, the process proceeds to step S307.
[0128]
In step S <b> 307, the CPU 21 executes a composite experience preference list creation process. That is, for each experience item T set in the experience item list of the experience favorable accumulation buffer 23E, the CPU 21 obtains an average value of the favorable sensitivity K set in the corresponding favorable sensitivity array 23G. For example, in the example of FIG. 27, the average value of favours 1-1 to 1-4 is obtained for the experience item T1, and the average value of positivities 2-1 to 2-8 is obtained for the experience item T2. Is required. Further, the CPU 21 rearranges the experience items T in descending order of the average value of the calculated preference K, and creates a composite experience preference list 23H as shown in FIG. Details of the processing in step S307 will be described later with reference to the flowchart in FIG.
[0129]
In this way, when the combined experience sensation list 23H is created, the process is terminated.
[0130]
Next, with reference to the flowchart of FIG. 29, the details of the composite experience preference list creation process in step S307 of FIG. 25 will be described. In step S501, the CPU 21 sets the length of the combined experience preference list 23H in the variable N, and sets 1 in the variable I representing the list length of the assigned portion. Next, in step S502, the CPU 21 sets a minimum favorable sensitivity limit value to av_min. Note that av_min may be set to 0 when there is no minimum favorable sensitivity limit value. In step S503, the CPU 21 sets the head pointer of the experience sensation accumulation buffer 23E (FIG. 27) to a variable K.
[0131]
In step S504, the CPU 21 determines whether the variable K is NULL. If the value of the variable K is not NULL, the process proceeds to step S505, and the CPU 21 calculates the average value of the favorableness of the favorableness array 23G associated with the experience item represented by the variable K. The details of this process are shown in the flowchart of FIG.
[0132]
That is, in step S531, the CPU 21 sets 0 for the total array length and the favorable value accumulated value. In step S532, the CPU 21 sets the first element of the array in the array scan pointer. For example, in the example of FIG. 27, the value of the array scanning pointer is set to the preference 1-1.
[0133]
In step S533, the CPU 21 adds 1 to the total array length, and adds the positivity value of the array element pointed to by the array scan pointer to the positivity accumulated value. In the present case, the value of the positive sensitivity 1-1 is added to the cumulative favorable value. Next, in step S534, the CPU 21 sets the next element of the array in the array scan pointer. In this case, the favorability 1-2 is set as the array scan pointer.
[0134]
Next, in step S535, the CPU 21 determines whether or not the next element exists in the array. In this case, since the next array exists, the process returns to step S533, and the CPU 21 adds 1 to the total array length again, and the positivity of the array element pointed to by the array scan pointer is added to the positivity accumulated value (in this case) , The value of positivity 1-2). In step S534, the CPU 21 sets the next element of the array in the array scan pointer. In the present case, the array of favorableness 1-3 is set as the array scanning pointer. In step S535, the CPU 21 determines whether or not the next element exists in the array. If there is, the process returns to step S533, and the subsequent processing is repeatedly executed.
[0135]
As described above, the same processing is repeated until it is determined in step S535 that the next element does not exist in the array.
[0136]
In this case, when the positivity values of sensitivities 1-1 to 1-4 are accumulated, NO is determined in step S535, and the process proceeds to step S536. The CPU 21 calculates the average from the following equation: Calculate likability.
[0137]
Average likability = Favorable cumulative value / total array length (9)
That is, in the case of the present example, a value obtained by dividing the cumulative addition value of the favourites 1-1 to 1-4 by the total array length of 4 is set as the average favorability. Then, this average favorable value is set in the variable ave.
[0138]
When the average favorable rate ave is calculated as described above, the process proceeds to step S506 in FIG. 29, and the CPU 21 determines that the average favorable rate ave calculated in step S505 is the minimum favorable sensitivity limit set in step S502. Determine if value is greater than av_min. If the average favorable sensitivity ave is equal to or smaller than the minimum favorable sensitivity limit value av_min, the experience item has no meaning to be processed, so the process proceeds to step S507, and the CPU 21 sets the variable K to the next pointer. Set the value. Then, the process proceeds to step S504, and the subsequent processing is repeatedly executed.
[0139]
In step S506, if it is determined that the average favorable value ave calculated in step S505 is greater than the minimum favorable sensitivity limit value av_min set in step S502, the process of registering the experience item in the composite experience favorable list 23H Is executed as follows. In other words, in step S508, the CPU 21 sequentially compares the favourites of the experience sensation element data E after the top of the composite experience sensation list 23H and before the I-th with the average favorite ave (combination experience sensation list). The average favorable sensitivity already registered in 23H is compared with the average favorable sensitivity ave calculated in this order), and the position of the experience favorable element data E that first becomes less than the average favorable sensitivity ave is set in the variable J. In step S509, the CPU 21 makes a pair of the Kth experience item of this time and the value of the average favorable sensitivity ave (as experience favorable element data) at the position immediately before the Jth, and puts it in the composite experience favorable feeling list 23H. insert. For example, the experience experience element data from the last I (N-1 when I is N) to the first Jth of the experience experience preference list 23H are moved backward one by one (to the smaller value of the preference value). After shifting, the Kth experience item of this time and the value of the average favorability ave are inserted as a pair (as experience favor element data).
[0140]
Next, the process proceeds to step S510, where it is determined whether or not the list length I of the assigned part of the combined experience preference list 23H is smaller than the entire list length N. If YES, the process proceeds to step S511, and the CPU 21 After incrementing the variable I by 1, the process proceeds to step S507, the value of the next pointer is set to the variable K, the process returns to step S504, and the subsequent processing is executed. That is, in this case, in FIG. 27, regarding the second experience item from the top, the same processing as in the case of the experience item 1 described above is executed.
[0141]
As described above, in step S510, when it is determined that the value of the list length I of the assigned part of the combined experience preference list 23H is equal to the value of the entire list length N, the process proceeds to step S512, and the CPU 21 , Av_min is set to the preference of the Nth experience preference element data. In other words, this prevents experience sensation element data having a likability equal to or smaller than that of the N-th experience sensation element data from being registered in the combined experience sensation list. Then, the process returns to step S507, the value of the next pointer is set in the variable K, and the processing after step S504 is repeatedly executed. The process is repeated until it is determined in step S504 that the variable K is NULL. If it is determined that the variable K is equal to NULL, the process completes the composite experience preference list creation in step S307 of FIG. The process proceeds to step S75 in FIG.
[0142]
As described above, the combined experience feeling list 23H (FIG. 27) is generated.
[0143]
In the process of step S303, the experience sensation element data E stored in the personal experience sensation record R of the compositing target member may be selected based on various conditions and set in the compositing target member experience sensation list 23F. it can. For example, only the experience sensation element data E posted within one month can be set in the compositing target member experience sensation list 23F. Based on this, a composite experience sensation list 23H is created. In this case, a member who does not hold the experience sensation element data E that satisfies this condition does not become a composition target member. In addition to the date and time, various conditions such as the experience location can be set as conditions for selecting the experience likability element data E, and a plurality of conditions can be set in combination.
[0144]
Next, details of the recommended experience (CD) list creation process in step S75 of FIG. 16 will be described. Based on the combined experience feeling list 23H created in step S74, the CPU 21 of the experience feeling information providing server 14 creates a recommended experience (CD) list 23I as shown in FIGS. Control the terminal 11 to transmit.
[0145]
The content of the recommended experience list 23I may be that the composite favorable feeling (average favorable sensitivity in FIG. 28) of the composite experience favorable feeling list 23H is used as it is as the recommendation degree of the corresponding experience item (CD title) (in this case) The recommended experience list 23I is a list similar to the composite experience favorable feeling list 23H shown in FIG. 28 (for example, a list as shown in FIG. 12), or does not include a recommendation level value and is simply a CD title. (In this case, the recommended experience list 23I is a list (for example, a list as shown in FIG. 14) from which the average favorable feeling of the combined experience favorable feeling list 23H shown in FIG. 28 is deleted).
[0146]
Further, the number of CD titles included in the recommended experience list 23I is limited to n (in this case, the recommended experience list 23I has n (n << L) (for example, a list as shown in FIG. 14), CD titles with a minimum recommendation level (average preference value is a predetermined threshold value or more) are included (in this case, recommendation) The experience list 23I is m from the top of the composite experience favourite list 23H shown in FIG. 28 (where m is the number of CD titles included in the composite experience favourite list 23H and the average likability value is greater than or equal to the threshold value. (For example, a list as shown in FIG. 13)).
[0147]
In addition, the characters of the experience items in the recommended experience list 23I can be emphasized corresponding to the degree of recommendation. Furthermore, when the recommended experience list 23I is not readable by humans, the information of the recommended experience list 23I can be output in some binary format.
[0148]
Also, a user introduction service can be provided based on the user list obtained as a result of the processing up to step S73 in FIG. 16 (the user buffer 23B in FIG. 23 or FIG. 24). For example, the information provided by the lover introduction service can include loyalty introduction information such as an email address. However, in this case, in order to protect the privacy, the member can limit the provision of privacy information. For example, it is possible to permit the privacy information to be provided to the lover only when there is a request from a member who is regarded as the lover in the specific experience that has been applied in advance. Actually, in the experience preference information providing server 14, for example, the privacy information provision restriction condition is stored in correspondence with the member's email address, and when the friend introduction list is created, it corresponds to the friend's email address. Only when the privacy information provision restriction conditions are confirmed and the conditions are satisfied, the privacy information is included in the lover introduction list.
[0149]
Furthermore, the above-mentioned privacy information provision restriction conditions can be set in combination with a plurality of conditions such as being a lover in a specific experience previously applied and being able to write Japanese. The privacy information provision restriction condition is provided by the administrator of the experience sensation information provision server 14 in which some conditions are selected by the member or freely selected by the member using the user interface. It can also be set.
[0150]
Next, I will introduce the associative experience. This is characterized in that the field used for creating the experience likability lists 23A and 23D to be referred to during the lover extraction process in step S73 of FIG. 16 is different from the field used for creating the composite experience list in step S74 of FIG. It becomes. For example, if a member who requested creation of a recommended experience list for music articles does not have a lot of experience sensation element data E for the music article experience, there is a possibility that a sufficient recommended experience (music article) list will not be created as it is. There is. However, if the member who requested the creation of the recommended experience list has a lot of experience feeling element data E for the CD experience, for example, first, the likes are extracted based on the experience feeling element data E for the CD experience, Furthermore, if a synthetic experience likability list is created based on the experience sensation element data E with respect to the music article experience of the extracted lovers (that is, the synthetic experience likability list can be created based on the associative experience). The member can get a full list of recommended experiences (music articles).
[0151]
FIG. 31 shows an example of a GUI when such a recommendation based on an associative experience is performed. As shown in the figure, in this example, a field where the user has already posted many experiences is input to the input unit 131. In addition, the input units 132 to 135 are configured to input information contents that the user wants to receive recommendations. The information input in the input fields 132 to 135 is the same information as the information input in the input fields 111 to 114 in FIG.
[0152]
When the user makes an input based on the GUI as shown in FIG. 31 and requests provision of associative experience information (presses the request button 136), the terminal 11 sends the experience preference information providing server 14 to FIG. Association request data as shown in FIG. This association request data is data corresponding to the input in the input fields 131 to 135 shown in the GUI of FIG. The preferred field corresponds to the data (“CD” in this example) input to the input unit 131 in FIG. 31, and the recommendation request is the data input to the input unit 132 (“book” in this example). It corresponds to. The minimum favorable sensitivity limit value av_min corresponds to the data input to the input unit 133. In this example, “recommendation only” is specified, so the minimum favorable sensitivity limit value av_min is set to 0.5. . The number limit corresponds to the input from the input unit 134, and the favorability corresponds to the input from the input unit 135.
[0153]
When the associative request data as shown in FIG. 32 is transmitted from the terminal 11, the CPU 21 of the experience preference information providing server 14 executes the process shown in the flowchart of FIG. 33. First, in step S601, the CPU 21 designates the member of the terminal 11 that has requested the provision of associative experience information as a member of interest. In step S602, the CPU 21 accesses the database designated in the favorite field field of the target member designated in step S601. For example, in the case of the example of FIG. 32, since the CD is designated as the favorite field, the CPU 21 stores the member personal experience feeling database 151 regarding the CD as shown in FIG. to access.
[0154]
In step S603, the CPU 21 executes a lover extraction process using the database accessed in step S602. This lover extraction process is the same as that described with reference to FIG. In other words, by this, the like-minded members who have similar preferences with respect to the member of interest and the CD title are extracted.
[0155]
Next, in step S604, the CPU 21 accesses the database specified in the recommended member's recommendation request field. In this case, as shown in FIG. 32, since “book” is designated in the recommendation request field, the CPU 21 selects a book out of the databases stored in the hard disk 27 as shown in FIG. The member personal experience favorable feeling database 152 is accessed. In step S605, the CPU 21 executes a combined experience likability list generation process. This composite experience preference list generation process is the same process as described with reference to the flowchart of FIG. 25 except that the experience item is not a CD title but a book title. That is, by this process, a combined experience likability list relating to the book title is created. In step S <b> 606, the CPU 21 generates a recommended experience list from the combined experience preference list and outputs it to the terminal 11.
[0156]
In this way, in this example, lovers who like the same CD as the member that the member of interest prefers are extracted, the book that the member likes is searched for, and the book becomes the member of interest as a recommended experience list. Provided.
[0157]
In this way, the recommended experience list 23I is generated based on the experience feeling element data E regarding the second field (recommendation request field) of the lovers who have similar preferences with respect to the first field (the field for searching for a friend). Created and provided to the member who requested it.
[0158]
As described above, since the personal experience sensation record R is updated by the experience sensation element data E posted by the member, preference information in all fields of the member can be collected, and the service (recommended experience list The scope of services provided) can be expanded.
[0159]
In the above, the case where the present invention provides a service related to the CD experience has been described as an example. However, the present invention is not limited to this and can be applied to the case where various services are provided.
[0160]
In this specification, the term “system” refers to an overall apparatus composed of a plurality of apparatuses and means.
[0161]
【The invention's effect】
  According to the present invention,Service information corresponding to interest and preference trends can be provided.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating an example of a network configuration of a server for providing experience-preferred information to which the present invention is applied.
FIG. 2 is a block diagram showing an example of the internal configuration of the experience sensation information providing server 14 of FIG. 1;
3 is a block diagram showing an example of the internal configuration of a terminal 11 in FIG. 1. FIG.
FIG. 4 is a diagram illustrating an example of a personal experience sensation record R;
FIG. 5 is a flowchart for explaining a post acceptance process of the experience favorable feeling information providing server 14 of FIG. 1;
6 is a diagram showing a display example of a GUI of the terminal 11 in FIG. 1. FIG.
7 is a diagram showing a display example of a GUI of the terminal 11 in FIG. 1. FIG.
8 is a diagram showing a display example of a GUI of the terminal 11 in FIG. 1. FIG.
9 is a diagram showing a display example of a message on the terminal 11 of FIG. 1. FIG.
10 is a flowchart for explaining processing of the experience sensation information providing server 14 of FIG. 1; FIG.
11 is a diagram showing a display example of a GUI of the terminal 11 of FIG.
12 is a diagram showing a display example of recommended experience information of the terminal 11 of FIG. 1. FIG.
13 is a diagram showing a display example of recommended experience information of the terminal 11 of FIG. 1. FIG.
14 is a diagram showing a display example of recommended experience information of the terminal 11 in FIG. 1. FIG.
FIG. 15 is a flowchart for explaining personal experience likability record update processing;
FIG. 16 is a flowchart for explaining recommended CD list creation processing;
FIG. 17 is a flowchart illustrating a detailed processing procedure of step S73 of FIG.
FIG. 18 is a diagram showing an example of an attention individual experience likability list 23A.
FIG. 19 is a diagram illustrating an example of a lover buffer 23B.
FIG. 20 is a diagram showing an example of a comparison target member experience preference list 23D.
FIG. 21 is a flowchart illustrating a detailed processing procedure of step S105 of FIG.
FIG. 22 is a diagram illustrating an example of a lookup table.
FIG. 23 is a diagram showing another example of the likes buffer 23B.
FIG. 24 is a diagram showing still another example of the likes buffer 23B.
FIG. 25 is a flowchart illustrating a detailed processing procedure of step S74 in FIG.
FIG. 26 is a flowchart illustrating a detailed processing procedure of step S304 in FIG.
FIG. 27 is a diagram illustrating an example of an experience favorable accumulation buffer 23E.
FIG. 28 is a diagram illustrating an example of a combined experience sensation list 23H.
FIG. 29 is a flowchart illustrating a detailed processing procedure of step S307 in FIG.
FIG. 30 is a flowchart illustrating a detailed processing procedure of step S505 in FIG. 29;
31 is a diagram showing a display example of a GUI of the terminal 11 of FIG.
32 is a diagram for explaining a format of data output from the terminal 11 of FIG. 1 corresponding to the display example of FIG. 31;
33 is a flowchart for explaining the operation of the experience-preferred information providing server 14 of FIG.
FIG. 34 is a diagram showing an example of a database.
[Explanation of symbols]
11 terminal, 12 base station, 13 telephone line, 14 experience information server, 21 CPU, 22 ROM, 23 RAM, 24 input unit, 25 display unit, 26 communication unit

Claims (13)

In an information processing device that selects information for a plurality of terminal devices,
Storage means for storing a personal information list having a plurality of personal information including experience information regarding information of each user using the plurality of terminal devices;
By comparing the experience information included in one personal information of interest with the experience information included in each personal information included in the personal information list, from among the individual information included in the personal information list, Personal information extracting means for extracting personal information highly relevant to the personal information of interest;
Selection means for selecting information for the terminal device corresponding to the personal information of interest based on the personal information extracted by the personal information extraction means;
Presentation control means for controlling the presentation of information selected by the selection means ,
The storage means is provided for each field having a plurality of items, and has a plurality of storage units for storing a personal information list for each field,
The personal information extraction means is a single storage in which the personal information of the user of the terminal device related to the first field input in the terminal device that requests selection of information among the plurality of storage units is stored. Based on a plurality of personal information stored in the section, extracting personal information highly relevant to the personal information of interest,
The selection unit is different from the one storage unit used by the personal information extraction unit among the plurality of storage units for personal information of the user of the terminal device corresponding to the personal information extracted by the personal information extraction unit. Information that selects from a plurality of personal information stored in another storage unit related to the second field input in the terminal device that requests selection of information, and selects information based on the specified personal information Processing equipment. The presentation control unit controls the presentation of information selected by the selection unit so as to display in a different format according to a condition specified by a user of a terminal device that requests the presentation of information. Information processing device. The information processing apparatus according to claim 1, wherein the presentation control unit controls presentation of information selected by the selection unit via a network. The information processing apparatus according to claim 1, wherein the experience information includes item information indicating an item of information. The information processing apparatus according to claim 4, wherein the selection unit selects information based on shareability between item information included in the personal information of interest and item information included in the identified personal information. The information processing apparatus according to claim 5, wherein the experience information includes item information indicating an item of information and likability information indicating likability with respect to the item of information indicated by the item information. The selection means includes a matching situation between the item information included in the personal information of interest and the item information included in the specified personal information, the preference information included in the personal information of interest, and the specified personal information. The information processing apparatus according to claim 6, wherein information is selected based on the likability information that matches the corresponding item information among the likability information included in the information. The information processing apparatus according to claim 1, further comprising personal information receiving means for receiving the personal information transmitted from the plurality of terminal devices. The selection means includes
By synthesizing a plurality of personal information identified to generate a combined personal information,
The information processing apparatus according to claim 1, wherein information for a terminal apparatus corresponding to the focused personal information is selected based on the synthesized personal information. Said selection means, the information to be selected based on the identified personal information, that limits to the predetermined number of information
The information processing apparatus according to 請 Motomeko 1. In an information processing method of an information processing apparatus configured by a computer that selects information for a plurality of terminal devices,
A storage step in which a hard disk of the computer stores a personal information list having a plurality of personal information including experience information regarding information of each user who uses the plurality of terminal devices;
By comparing the experience information included in one personal information of interest with the experience information included in each personal information included in the personal information list, from among the individual information included in the personal information list, A personal information extraction step in which the CPU of the computer extracts personal information highly relevant to the personal information of interest;
Based on the personal information extracted in the personal information extraction step, a selection step in which the CPU of the computer selects information for a terminal device corresponding to the personal information of interest;
The presentation of the information selected by the selecting step, seen including a presentation control step of the CPU of the computer control,
The storage step is provided for each field having a plurality of items, and a personal information list for each field is stored in a plurality of storage units provided in the hard disk of the computer,
In the personal information extraction step, one storage in which the personal information of the user of the terminal device related to the first field input in the terminal device that requests selection of information among the plurality of storage units is stored. Based on a plurality of personal information stored in the section, extracting personal information highly relevant to the personal information of interest,
The selection step is different from the one storage unit that uses the personal information of the user of the terminal device corresponding to the personal information extracted in the personal information extraction step, among the plurality of storage units, in the personal information extraction step. Information that selects from a plurality of personal information stored in another storage unit related to the second field input in the terminal device that requests selection of information and selects information based on the specified personal information Processing method. In a recording medium on which a computer-readable program for selecting information for a plurality of terminal devices is recorded,
Storing a personal information list having a plurality of personal information including experience information regarding information of each user using the plurality of terminal devices in a hard disk of the computer;
The personal information list is obtained by comparing the experience information included in the personal information of interest with the experience information included in the personal information included in the personal information list for one personal information of interest. A personal information extraction step for causing the CPU of the computer to extract personal information highly related to the personal information of interest from each personal information included in the computer;
Based on the personal information extracted in the personal information extraction step, a selection step for causing the CPU of the computer to select information for a terminal device corresponding to the personal information of interest;
The presentation of the information selected by the selecting step, seen including a presentation control step of controlling the CPU of the computer,
The storage step is provided for each field having a plurality of items, and a personal information list for each field is stored in a plurality of storage units provided in the hard disk of the computer,
In the personal information extraction step, one storage in which the personal information of the user of the terminal device related to the first field input in the terminal device that requests selection of information among the plurality of storage units is stored. Based on a plurality of personal information stored in the section, extracting personal information highly relevant to the personal information of interest,
The selection step is different from the one storage unit used in the personal information extraction step among the plurality of storage units, the personal information of the user of the terminal device corresponding to the personal information extracted in the personal information extraction step. , Selecting from a plurality of personal information stored in another storage unit related to the second field input in the terminal device that requests selection of information, and selecting information based on the specified personal information
A recording medium on which a program for causing a computer to execute processing is recorded. In an information processing system comprising a plurality of terminal devices and an information processing device that supplies service information to the plurality of terminal devices,
The plurality of terminal devices are:
Input means for inputting experience information related to the service information and request information for requesting the service information from the information processing apparatus;
First communication means for transmitting the experience information and the request information to the information processing apparatus and receiving the service information from the information processing apparatus via a network;
Display means for displaying the service information received by the first communication means,
The information processing apparatus includes:
Receiving the experience information and the request information from the terminal device via the network;
Second communication means for transmitting the service information to the plurality of terminal devices;
Storage means for storing a personal information list having a plurality of pieces of personal information including the experience information relating to the service information transmitted from the first communication means of the plurality of terminal devices received by the second communication means; ,
Attention personal information extraction means for extracting from the storage means the experience information included in attention personal information that is personal information regarding the terminal device that has transmitted the request information;
Each personal information included in the personal information list is compared by comparing the experience information included in the personal information of interest with the experience information included in the personal information list included in the personal information list stored in the storage unit. Related personal information extracting means for extracting personal information highly relevant to the noted personal information from
And selecting means for on the basis of the personal information extracted by related personal information extraction means, for selecting the service information to the corresponding terminal apparatus and the target personal information,
Presentation control means for controlling the presentation of the service information selected by the selection means ,
The storage means is provided for each field having a plurality of items, and has a plurality of storage units for storing a personal information list for each field,
The related personal information extracting means stores one of the plurality of storage units that stores the personal information of the user of the terminal device related to the first field input in the terminal device that requests selection of information. Based on a plurality of personal information stored in the storage unit, to extract personal information highly relevant to the noted personal information,
The selection means includes the one storage unit that uses the personal information of the user of the terminal device corresponding to the personal information extracted by the related personal information extraction means, among the plurality of storage units, by the related personal information extraction means. Different from the above, specifying from a plurality of personal information stored in another storage unit related to the second field input in the terminal device that requests selection of information, and selecting information based on the specified personal information an information processing system to be.

Images